The BE breakdown video is great ?
I'm working on a PHD where this effect plays a role. May I use your video in my thesis ?
Just speculating, is this a MESA part ?
Just speculating, is this a MESA part ?
Sounds reasonable! :-+
... just a question - is the 2N2222A also emitting light in BE breakdown ?
regards
Wolfgang
Is it possible to check tip3055 and tip2955 from onsemi and st? I'm willing to sponsor this. They are in plastic enclosure, to-264 or to-247.
I have uploaded a new row of pictures including the current values:
https://www.richis-lab.de/Bipolar02.htm (https://www.richis-lab.de/Bipolar02.htm)
One interesting effect is the positive temperature coefficient of the breakdown voltage (Z-diode with Vbr=10V). If you connect a voltage at B-E that is just high enough to let it break down and then increase the current the B-E-junction suddenly gets non conductive again. => Physics! :-+ ;D
Very interesting results with the optical output from the reversed breakdown EB junction!! 8)
Also like the method you've developed to de-encapsulate ICs based upon "burning" off the epoxy, very clever :-+
An interesting effect on some bipolar transistors when used with EB junction reversed biased in this type behavior where the plot of voltage vs. current has a region of negative slope when entering breakdown, or dV/dI is negative, thus negative resistance. This is the type of behavior of a avalanche or tunnel diode.
...
The late Jim Williams (brilliant analog engineer, RIP) discussed at the ISSCC awhile back (I had a brief private discussion with him) how he designed a cheap voltage reference using a small NPN transistor with the EB junction operated in breakdown which produces a positive TC, then the base was left open and the collector grounded. This would forward bias the base to collector diode would produce the negative TC compensative the positive TC of the reversed breakdown of the EB junction. This went into production only to find that the composite transistor based reference was oscillating using the decoupling capacitor and resistor bias!!
Best,
I have uploaded a new row of pictures including the current values:
https://www.richis-lab.de/Bipolar02.htm (https://www.richis-lab.de/Bipolar02.htm)
You had a private discussion with Jim Williams? Very cool! 8)
BTW what current did you run thru the LTZ1000 to see the optical output? Would really like to get an LTZ1000, but they've become very expensive lately :-\
... just a question - is the 2N2222A also emitting light in BE breakdown ?
... just a question - is the 2N2222A also emitting light in BE breakdown ?
Coincidentally I had a 2N2222A in my inbox. ;D
(https://www.richis-lab.de/images/transistoren/04x05.jpg)
As expected the BE-junction glows too.
I then killed the 2N2222A and made a short video. >:D
(https://www.richis-lab.de/images/transistoren/04x09.png)
You can identify the route of destruction:
(https://www.richis-lab.de/images/transistoren/04x09.jpg)
My Interpretation:
1) Firstbreakdowndestruction of the BE-junction.
2) Base-Electrode melts and cuts the current.
3) Second destruction of the BE-junction.
4) Base-metal melts further to the bondpad.
5) Last connection is disrupted with a bright arc.
Interesting... :popcorn:
More pictures here:
https://www.richis-lab.de/Bipolar04.htm (https://www.richis-lab.de/Bipolar04.htm)
:popcorn:
- the video shows a very fast current increase. Did you run this with a current source ? Or is just the video too fast ?
- In my transistors, I see a noise curve very explainable by your KD503 video. First, only a few hotspots light up,
then more and more, until the whole area is bright. So noise *falls* with rising current. The 2N2222 seems to start almost immediately. Why ?
The Zener range is the same (> 7V), so it must be an avalanche mechanism.
- Do you want noise plots ? In case the 2N2222A behaves the same, they could be of interest.
Thanks again ! If you dont mind, I really want to give you an honourable mention in a paper I'm writing.
I wonder if the glow pattern varies with charge state on the surface passivation (which is likely silica glass). This would be hard to test; perhaps exposed metal surfaces could be insulated with an insulating film, then a conductive liquid (e.g. salt water) applied to control the surface electric field? (Might need hundreds of volts, since the insulation will be so thick at this point.)
Tim
Why nobody put a photodiode inside to protect expensive power transistors?
Why nobody put a photodiode inside to protect expensive power transistors?
Why nobody put a photodiode inside to protect expensive power transistors?
Während die Z-Diode leitet arbeitet sie zumindest zum Teil im Lawinendurchbruch. Wie bei den Versuchen mit den 2N3055-Transistoren ist dabei im Bereich der Sperrschicht ein Leuchten zu erkennen. Rekombinieren Ladungsträger in einem Siliziumhalbleiter, so emittieren sie üblicherweise kein Licht im sichbaren Bereich. Bei einem Lawinendurchbruch erfolgen allerdings relativ unkontrollierte Ionisierungen im Kristallgitter, die unter anderem auch sichtbares Licht erzeugen.My German is rather poor, so I used google translator
While the Zener diode is conducting, it works at least in part in the avalanche breakdown. As in the experiments with the 2N3055 transistors, a glow can be seen in the area of the junction. If charge carriers recombine in a silicon semiconductor, they usually do not emit light in the visible range. In the event of an avalanche breakdown, however, relatively uncontrolled ionizations occur in the crystal lattice, which among other things also generate visible light.
What magnitude? Ic should be at least ~nA to start with; I would guess uA is reasonable to expect here?
I'm not sure whether this was debated here, but indeed the principle behind the glow is a bit peculiar one. I found relevant part of a book "Handbook of Silicon Photonics" here
https://books.google.sk/books?id=6zjNBQAAQBAJ&pg=PA347&lpg=PA347#v=onepage&q&f=false (https://books.google.sk/books?id=6zjNBQAAQBAJ&pg=PA347&lpg=PA347#v=onepage&q&f=false)
In a case the link above becomes dead, attached is excerpt with relevant part.
If you have access to a metal lathe, you need to build this:
...
I wonder why using two dies? As I see, both dies are in parallel without any balansing resistors. They must be very well matched, and, from the secondary breakdown pic, it looks like they are.
I also noticed a few red spots outside where they shouldn't be. I wonder what's that. Like, the one on the right die near the bottom pad.
I wonder why using two dies? As I see, both dies are in parallel without any balansing resistors. They must be very well matched, and, from the secondary breakdown pic, it looks like they are.
I also noticed a few red spots outside where they shouldn't be. I wonder what's that. Like, the one on the right die near the bottom pad.
Not reeeeally... I've seen single MOSFET and IGBT dies bigger than the inside area of a TO-3. More likely it was contemporary yields.
Don't know what the largest BJT die is, these days; might not even be one as large, just because there's so little demand for them.
Tim
I wonder why using two dies? As I see, both dies are in parallel without any balansing resistors. They must be very well matched, and, from the secondary breakdown pic, it looks like they are.
I also noticed a few red spots outside where they shouldn't be. I wonder what's that. Like, the one on the right die near the bottom pad.
We could drill out the remaining pin hole, and get a cheap matched pair transistor? :)
Not reeeeally... I've seen single MOSFET and IGBT dies bigger than the inside area of a TO-3. More likely it was contemporary yields.
Don't know what the largest BJT die is, these days; might not even be one as large, just because there's so little demand for them.
Tim
Same problem with MOSFETs. IXYS uses graded threshold voltages to tackle this problem (high threshold in the center) for their linear MOSFETSs.
All others - same story, with hotspotting in linear mode. See NASA, Spirito Effect.
Not reeeeally... I've seen single MOSFET and IGBT dies bigger than the inside area of a TO-3. More likely it was contemporary yields.
Don't know what the largest BJT die is, these days; might not even be one as large, just because there's so little demand for them.
Tim
Same problem with MOSFETs. IXYS uses graded threshold voltages to tackle this problem (high threshold in the center) for their linear MOSFETSs.
All others - same story, with hotspotting in linear mode. See NASA, Spirito Effect.
The "reeeeally" being, most transistors that size are made for switching, so may have awful SOAs. The power dissipation is there, no contest, just doing it at voltage is harder.
That said, many newer MOSFETs, and even some IGBTs, are specified with DC SOA. However they've approached it -- graded threshold, tempco hackery, ballasting*, whatever -- it's done the trick.
*Probably not that, because source/emitter degeneration would severely eat into the saturated performance.
Tim
I don't believe I've ever seen two dice paralled as in the BUX22 above. Looking at a data sheet, I don't see anything calling that out. I expect the dice are matched before packaging. Any thoughts on the characteristics that make this possible? I could see that emitter ballasting would help.
The BD522 looks like it's got eyes, LOL! :D
You connect the base over the whole die through perforations of the emitter. That gives you a better current Distribution and that leads to lower saturation voltage and second breakdown appears later.
Here you have more perforated emitter:
https://www.richis-lab.de/2SC2922.htm (https://www.richis-lab.de/2SC2922.htm)
https://www.richis-lab.de/Bipolar08.htm (https://www.richis-lab.de/Bipolar08.htm)
You connect the base over the whole die through perforations of the emitter. That gives you a better current Distribution and that leads to lower saturation voltage and second breakdown appears later.
Here you have more perforated emitter:
https://www.richis-lab.de/2SC2922.htm (https://www.richis-lab.de/2SC2922.htm)
https://www.richis-lab.de/Bipolar08.htm (https://www.richis-lab.de/Bipolar08.htm)
do you by accident have an 2N2857 at hand ?
I would be curious about this one.
Thank you @Noopy, those articles taxed my German vocabulary to the limit, aber es ist etwas gut!
Do you know why integrated SMART-Highside-Driver have low clamping voltages?
There is a parasitic bipolar transistor between the power transistor and the barrier around the logic part that is connected to the ground potential. You don´t need very much voltage to break the collector-emitter-line and kill the part.
The datasheet explains that the overcurrent detection is done with the temperature meassurement (this small satellite) but it seems that there are two small MOSFETs. Theses small MOSFETs can be used to sense the current through the VN02H...
Do you know why integrated SMART-Highside-Driver have low clamping voltages?
There is a parasitic bipolar transistor between the power transistor and the barrier around the logic part that is connected to the ground potential. You don´t need very much voltage to break the collector-emitter-line and kill the part.
I assume the dielectricly isolated process used for DMOS would be too expensive in both fabrication and area.
The datasheet explains that the overcurrent detection is done with the temperature meassurement (this small satellite) but it seems that there are two small MOSFETs. Theses small MOSFETs can be used to sense the current through the VN02H...
That is a very common technique for CMOS power devices and 4 pin power MOSFETs which allow monitoring the current. A few cells of the power MOSFET have a separate drain connection brought out and the ratio of areas determines the fraction of the drain current which is detected.
An LM395 die shot would be interesting and I think I have seen it somewhere. As I recall, it was very similar to the LM317 and might have been the same die with a different metalization.
Do you know why integrated SMART-Highside-Driver have low clamping voltages?
There is a parasitic bipolar transistor between the power transistor and the barrier around the logic part that is connected to the ground potential. You don´t need very much voltage to break the collector-emitter-line and kill the part.
I assume the dielectricly isolated process used for DMOS would be too expensive in both fabrication and area.
And with dielectric isolation you can´t build the more powerful vertical DMOS, can you?
A classical mesa-design but OnSemi uses it still today for the TIP2955:
https://www.richis-lab.de/Bipolar11.htm (https://www.richis-lab.de/Bipolar11.htm)
A classical mesa-design but OnSemi uses it still today for the TIP2955:
https://www.richis-lab.de/Bipolar11.htm (https://www.richis-lab.de/Bipolar11.htm)
I was under the impression that the mesa process was long discontinued, except maybe for premium parts, because it is too expensive due to the time it takes to manufacturer.
Rough structures gives you a lot of leackage so they etched some traces to get smooth junction edges.
Rough structures gives you a lot of leackage so they etched some traces to get smooth junction edges.
Like high voltage diodes, I thought the outer ring structure in thyristors was to prevent high voltage breakdown.
Rough structures gives you a lot of leackage so they etched some traces to get smooth junction edges.
Like high voltage diodes, I thought the outer ring structure in thyristors was to prevent high voltage breakdown.
I think we are talking about the same mechanism:
The amount of leakage current is depending of the voltage.
More voltage, more leakage, triggering the thyristor... :-BROKE
Time dependent false triggering in a thyristor is suppressed with metalization between the base and emitter of each transistor. "Sensitive gate" SCRs are sensitive because they lack this. An external resistor is less effective because of the distributed nature of the spreading resistance; the thyristor could be triggered in an area where there is too much resistance to the gate connection which is a very bad situation.
I was referring to the guard rings applied to semiconductor junctions including diodes and transistors which increase breakdown voltage.
It's intriguing to see so many wire bonds inside of a 2GHz part, thinking here about parasitic inductance and stray capacitance. ???
I would guess, for the technology of the day, they did the best with what they had: they probably found that etching the sidewalls, rather than leaving them open on the primary surface, simply gave better results. The device might not be very tolerant of high voltage or rate stresses (avalanche and pulse operation?), but also maybe it was low enough voltage that it worked out okay.
This fellow built a 3 phase linear amplifier using similar devices: http://wunderkis.de/pwramp3/index.html (http://wunderkis.de/pwramp3/index.html)That's me, by the way.
I toyed with the idea of making an electronic load with one of modules I have.I'd recommend against that:
Ah, the capt himself. Nice job on the driver and the write up is very clear - thank you!
Yes, the DC SOAs on the modules I have are not so good - a Pdmax of 200 W for a devices with a BVCEO of 500 V and an ICMAX of 100 A, clearly optimized for switching. Their instability was a bit of a surprise as the servo drive they are from had 10 to 30 cm long wires and bus bars all over the place carrying drive signals and switched currents. However, because it was a PWM, the devices weren't in the linear regime long enough to encounter trouble from oscillation. After encountering the oscillation and reading the write up, it's back to the old plan. Too bad, because the module be simple to mount on a heat sink and wire up.
Try linear mosfets (IXYS). They are made for electronic loads.
Try linear mosfets (IXYS). They are made for electronic loads.
The problem is that unless you need maximum power in fewer packages, bipolar transistors just cost less for a given die area and power dissipation is proportional to die area. So linear MOSFETs come with a double price premium which might even make lateral MOSFETs price competitive.
Please excuse my ignorance, where is the second bjt located? Is it two big pads on the left on the close up shot?
There is a second transistor on the die probably to test the die. Interesting... Why didn´t they do the testing with the 2N2857 transistor itself? :-//
There is a second transistor on the die probably to test the die. Interesting... Why didn´t they do the testing with the 2N2857 transistor itself? :-//
Could the tests be destructive like base-emitter breakdown voltage?
How do they pick and place such small dies? Or even cut them...
Those metal can packages... I have an urge to buy just for the sake of owning it.
PS I have Russian МП-42 (MP-42) somewhere, an old germanium transistor. I wanted to put it into use, but may be I should crack it open and see what's inside :). Can't find it atm, must be hiding from me....
There is a second transistor on the die probably to test the die. Interesting... Why didn´t they do the testing with the 2N2857 transistor itself? :-//
Could the tests be destructive like base-emitter breakdown voltage?
Never heard of such testing but that would explain the second transistor...
How do they pick and place such small dies? Or even cut them...Back in my times(30+years ago) I've opened quite a few from the MP38-42/P401-3 series; first, they are all Ge transistors; second, all of the structures are visible with the naked eye(die connection:1.00X0.01 mm Al or similar flat); third , don't blame my memory if there's something totally diferent inside; I just try to remember best of my impressions.
Those metal can packages... I have an urge to buy just for the sake of owning it.
PS I have Russian МП-42 (MP-42) somewhere, an old germanium transistor. I wanted to put it into use, but may be I should crack it open and see what's inside :). Can't find it atm, must be hiding from me....
Back in my times(30+years ago) I've opened quite a few from the MP38-42/P401-3 series; first, they are all Ge transistors; second, all of the structures are visible with the naked eye(die connection:1.00X0.01 mm Al or similar flat); third , don't blame my memory if there's something totally diferent inside; I just try to remember best of my impressions.
These two transistors have the same collector ;)
Also 2nd breakdown probably doesn't emit any visible light, so you'd need a high framerate IR camera at that!
Tim
That being said, you could still try basic avalanche breakdown of the BC junction. Just need a resistor-limited HV generator.
Perhaps that´s because most of the BC junction is covered under "a lot" of silicon. :-//That's possible, although usually the edges of the junction reach the surface.
If the glow persists in Vceo or Vces mode, I wonder if it has any physical significance -- sufficient light output could perhaps cause the junction to ionize much sooner, and more broadly, than it would due to carrier diffusion alone!
Another factor is beta: at 10mA collector current, only some 0.1mA may be flowing by means of breakdown and the rest because of the normal current gain of the transistor.
(https://www.richis-lab.de/images/Transistoren/24x13.jpg)
Hm... Somehow I managed to kill the 2N2857...
Unfortunatelly I can´t say what went wrong. I was in a hurry. That´s never a good thing. :-BROKE :-//
That´s not "medium" it´s more "well done". ;D
@Wolfgang: SMU and cheap are two words that don´t match. ;)
I still haven´t set up my HP-supply... :-\
I proudly did once a demonstration about my latest tool purchase, a brand new Rigol DP832 power source, and how great it is, and how it can serve as a voltage source, or as a current source, upon wish.
And to prove my point, I set the current to 20mA and confidently hooked up a LED to the wires. :-+
The LED flashed then instantly died with a violent pop sound, I jumped back, and my friend burst into laughter. :-DD
Those 20mA were enough to charge a few thousands uF of output capacitors up to 32V, then all the energy rushed into the LED, limited only by the ESR and the wires. In DP832, the output filtering capacitors are located at the end side, in direct contact with the output terminals. ;D
[...]
Those 20mA were enough to charge a few thousands uF of output capacitors up to 32V, then all the energy rushed into the LED, limited only by the ESR and the wires. In DP832, the output filtering capacitors are located at the end side, in direct contact with the output terminals. ;D
Those 20mA were enough to charge a few thousands uF of output capacitors up to 32V, then all the energy rushed into the LED, limited only by the ESR and the wires. In DP832, the output filtering capacitors are located at the end side, in direct contact with the output terminals. ;D
Interesting. Mine (DP832) ramps up beautifully, no inrush or spikes whatsoever on all 3CH, smooth as silk and stable. Recently had it calibrated also so add in that now impressive accuracy, its incredible now.
Anybody knows more about the 2 spheres formed at the end of the melted bonded wires from the last transistor's pics? Are they usually so round and alike looking?
If you don't have an SMU, a bunch simple resistors should do the job.
Use the lab supply in voltage mode, roughly calculate the required resistance and put the resistor in series to the DUT. Measure the actural current with your multimeter and adjust the output voltage to set the desired current.
Disadvantage: for covering a larger range of currents, one has to switch resistors.
Your typical old-style curve tracer (e.g. Tek 576) does it this way. Pretty simple and effective. The modern ones have multiple digitally controlled SMUs to achieve the same results.
Today I have a MJL21193 for you (250V/16A/30A):
[...]
(https://www.richis-lab.de/images/transistoren/23x02.jpg)
The die is quite big: 3,64mm x 3,54mm
Interesting with that old 2N3055, it seems technology has moved on and we are making much "cleaner" devices nowadays?
FERD diode please...
I can put it on my list. :-+
I didn´t test the switching speed but the TIP3055 is one of a batch that exe gave to me to take pictures. I think he did some measurements.
exe?
Hmm, do you know what h_fe was under that condition?
That should be the cutoff point, where it transitions from flat h_fe (from down to DC) to asymptotic (constant fT) behavior. Which means fT = h_fe * Fc.
Tim
Concerning the metal layer over the pn junction, Could it be to spread the heat?
Wow, those igbt and diode are quite large.It must be huge when
Wow, those igbt and diode are quite large.It must be huge when
IC25 TC = 25°C (Limited by Leads) 75 A
So bond wires are weaker than die itself >:D
Some of these pictures are quite beautiful and wouldn't look out of place at an art exhibition, LOL! :D
ft of 5kHz...
Why did they use a whole round crystal for transistors back then?
Is it some property of germanium that it can't be processed like silicon on bigger crystal and then cut to dies or just wasn't the technology to produce bigger crystal back then?
It will not explicitly answer your questions, but will let you make an idea about the zeitgeist of the electronic industry back then.It kinda explains it
Next "merchandise project" is a LTZ1000 coffee pot. :D
What's the purpose of the black goo around the dies?
Weird, it's common-gate?
Wait... are those not actually the can? And if they were, there wouldn't be any need for epoxy in the bottom. Could those be... gate pins, coined flat for mounting the dies on? Then the epoxy underfill is not so much to insulate (the hermetic seal should be fine) as to control vibration (so it's cemented down).
Hm, is the can floating, then?
Weird, it's common-gate?
Today I have a Dual-J-FET for you, the DN1682
I'm surprised, I expected to see two weirdly-interleaved fets on a single die.
Sometimes I go up to 1,5A
And its an interesting start problem for a computerized PSU / photography project. >:D
BTW, my power supply can do pulses, I just thought I'll be quick-enough to do it manually. I was wrong. At 10A there was really little time to do the shot. I think it was about a second or less before brightness dropped, then sparks showed up.
power dissipation = voltage drop · current
Still the die is potted with some red silicone.
Quite a big die for a TO220-package: 5,8mm x 4,3mm 8)
A lot of tiny MOSFETs.
Power MOSFETs brought on development of TO-220 packages which could hold larger dies, and eventually larger TO-220 style packages. Later TO-220 packages had a lead frame which barely fit within the encapsulation.
Since the manufacturer is Siliconix, that makes it MOSPOWER!
https://archive.org/details/bitsavers_siliconixdixMOSPOWERApplications_38092918
An International Rectifier part would be a HEXFET with a hexagonal arrangement of cells. Do they still make these? I blew up enough of them.
But really? KD501 at 1A? Those things can to 10A continues safely !
I thought the inside of the TO3s were dark cold places, indeed wrong lit by the current :D I like it.
If you ever come across a tesla germanium transistor such as 7NU74 73 or 6NU Take a look at those. I have milled one 6NU74 (because some sadistic murderer broke its legs off) and it turned out to be a metal capsule in which the semiconducotr was located. The metal "can" was split in 3 parts acting as electrodes I guess. Oh and It also had a rattly absorband inside. I rock a pair of 7NU74s in a AB amp and they are very interesting :)
can you share video link
How much did the transistors cost back in the day?
I also heard PNP transistors were more expensive than NPN.
FERD diode please...
I have a FRED in stock.
I have a FRED in stock.
You mean FERD, right? I'm in great anticipation! There is very little info on these beasts.
While searching for info, it seems there a newer generation called cc-ferd: https://www.semanticscholar.org/paper/A-New-Generation-of-Power-Diode%3A-Charge-Coupled-Lee-Ngwan/724f82fd90cbea5b81bb116e646b7105117ca701 (https://www.semanticscholar.org/paper/A-New-Generation-of-Power-Diode%3A-Charge-Coupled-Lee-Ngwan/724f82fd90cbea5b81bb116e646b7105117ca701) . I'm trying to find a full paper. Although, looking at research papers, there seems to be many ideas, but few make into silicon.
UP: found pdf scholar.google.com
Ha, the gate connection for a JFET is through the substrate but that is also a point of failure so they added a bond wire in parallel.
Ha, the gate connection for a JFET is through the substrate but that is also a point of failure so they added a bond wire in parallel.
In addition the bond wire reduces the gate impedance.
Ha, the gate connection for a JFET is through the substrate but that is also a point of failure so they added a bond wire in parallel.
In addition the bond wire reduces the gate impedance.
The connection through the substrate is much lower impedance than the bond wire, which also goes to the substrate.
The problem is that the substrate connection through the bottom of the die is difficult to make reliably. There are examples of this in the past from National and Tektronix where the substrate connection became intermittent with temperature but the transistor still worked because of capacitive coupling, except of course when it did not.
Ha, the gate connection for a JFET is through the substrate but that is also a point of failure so they added a bond wire in parallel.
In addition the bond wire reduces the gate impedance.
The connection through the substrate is much lower impedance than the bond wire, which also goes to the substrate.
The problem is that the substrate connection through the bottom of the die is difficult to make reliably. There are examples of this in the past from National and Tektronix where the substrate connection became intermittent with temperature but the transistor still worked because of capacitive coupling, except of course when it did not.
Are you sure? I thought silicon has always more resistance than metal? OK, the die is much thicker and shorter than the bondwire but nevertheless...
It's hard to believe they bonded a wire just in case the substrate connection fails. The behaviour would change all the same.
For a JFET the substrate connection goes to the gate so any series resistance is inconsequential, unless it becomes very high. And even when open, many circuits will still work because of capacitive coupling.
Here you can see this is in fact a transistor's C-B diode that is used for temperature sensing. It would be nice to find out which transistor could this be -roughly. It is possible that this is somewhat a custom part made by 3rd party vendor, but then why would they hassle with transistor structure when only a diode is needed?
Obviously the very precise pieces are heavily binned so I don't think there is a free meal here, but still it would be nice to get a rough idea on the source.
Could you please remind me what are those rings around the transistor?
Philips MPSA56, a pnp transistor.
Philips MPSA56, a pnp transistor.
On Semiconductor makes it also and isn't MPS or MPSA a Motorola prefix?
It is so similar to the 2N4403 that I wonder why it even exists. It is slightly slower but slightly higher voltage so I wonder if it is just a graded 2N4403.
It is so similar to the 2N4403 that I wonder why it even exists.
There are also MPSA transistors built by Philips. The one how gave me these transistors stated it is a Philips... :-//
Can it be a "European" equivalent of 2n4403? Like 2n3904 and bc548. Anyway, I'm pretty sure there are more part numbers than design variations. I bought BD135G, BD137G and BD139G (don't remember the vendor, there are several companiesproducingsupplying them) and measured them with transistor tester and with siggen for gain vs freq. They all had very close gain and base drop voltage, which I interpret that they are the same design.
There are also MPSA transistors built by Philips. The one who gave me these transistors stated it is a Philips... :-//
I know others make MPS and MPSA prefix transistors. I just thought that prefix originated with Motorola and others are second sources.
(https://www.richis-lab.de/images/Transistoren/71x01.jpg)
(https://www.richis-lab.de/images/Transistoren/71x01.jpg)
Today I´m pretty sure that´s a conterfeit part:
- 2005 Motorola was already ON Semiconductor.
- In the "Motorola Semiconductor Master Selection Guide 1994" there is already no TO-66 package left.
- Up to now I found such a white potting just in the conterfeit BUX22 (https://www.richis-lab.de/Bipolar09.htm (https://www.richis-lab.de/Bipolar09.htm)) and in the Inchange 3DD15D (https://www.richis-lab.de/Bipolar05.htm (https://www.richis-lab.de/Bipolar05.htm))
- The strange additional "ring" between base and emitter was also only found in the conterfeit BUX22 and in the Inchange 3DD15D.
>:D
Siemens BC239C
The die is 0,35mm x 0,35mm.
Well I don´t know... :-//
So like most brittle materials (glass, ceramic)Well I don´t know... :-//
I vaguely remember, the wafer gets scribed along the lines between the dice, then attached to some elastic material and pulled apart to separate the dice.
Perhaps there was also some binning.Yup, this was likely the case. Legend has it that Tesla did not gave great process yields or repeatability and they did a lot of binning for semiconductors.
Thanks for the great work!
Perhaps there was also some binning.Yup, this was likely the case. Legend has it that Tesla did not gave great process yields or repeatability and they did a lot of binning for semiconductors.
An interesting analysis would be a silicon capacitor ( https://www.mouser.sk/Passive-Components/Capacitors/Silicon-RF-Capacitors-Thin-Film/_/N-5g95 (https://www.mouser.sk/Passive-Components/Capacitors/Silicon-RF-Capacitors-Thin-Film/_/N-5g95) ). Some seem to have far more advanced structures on them :)
So who's hiding inside that transistor, then? :D
(https://www.richis-lab.de/images/Transistoren/75x01.jpg)
https://www.richis-lab.de/Bipolar46.htm (https://www.richis-lab.de/Bipolar46.htm)
Yes, I did wonder if you had found the packaging different / harder to remove. :)
I think you may find that the 'E' stands for E-line. ;) This family dates back to the Ferranti E-Line series.
They use a Silicone rather than Epoxy packaging material, allowing higher junction temperatures and dissipations than TO92. They were used a lot in Mil spec equipment.
...
Higher C-B leakage I think is a factor in Vcbo vs. Vceo breakdown, since the leakage is effectively multiplied; I'm not sure what all physics is really at work, as the ratio can be quite significant (as in this case); other times it's much more modest.
FWIW, I'd love to see microscopy of a bare die under such operation -- but I'm also afraid that there would be so little light emitted (due to the very low duty cycle, < 0.1% is typical) that it might not show at all.
(The usual setup is: a large pullup resistor to charge the collector, typically 10-100k to +100V or thereabouts; a B-E resistor, typically 4.7k or thereabouts, depending on type; and some kind of load, at least a few pF from C-E but also including an output coupling network like a 50 ohm transmission line or whatever. 100-120V is enough to cause 2N3904 to break down in this way; 2N2369 needs less, 60-80V I think; it even works for high voltage types, but because breakdown occurs as a narrow filament, power transistors are essentially dead weight for this -- a 1500V 10A transistor fails with only a few 100s of pF load, hardly more than a 300V 100mA type can handle.)
I have a question. Current western semiconductor companies such as onsemi, vishay, diodes, infineon, nexperia and rohm, do they still produce bjts by themselves, or they just relabel Chinese parts? Asking because bjts are so cheap, how they can make money from them...
Wouldn´t it be possible to test the CE-breakdown in a constant manner? If we use a low voltage transistor the power dissipation is perhaps low enough to take pictures for some seconds.
Wouldn´t it be possible to test the CE-breakdown in a constant manner? If we use a low voltage transistor the power dissipation is perhaps low enough to take pictures for some seconds.
Some amazing pictures there, @Noopy, have you got some new equipment or techniques you are playing with? Or do you just keep getting better? :D
Oh wow weird, is that... two, three thicknesses of metallization?
I've always wondered if they do things with metallization thickness but never saw much evidence for it (also, it's extra masks, why bother..?), sure looks like they did that here though. Would the thinner first layer (particularly for the emitter) serve as ballasting?
patienceaka focus stacking :P
patienceaka focus stacking :P
Although pnp transistors have often worse specifications than npn transistors the CDIL BC560C has the same edge length as the CDIL BC550C.
Although pnp transistors have often worse specifications than npn transistors the CDIL BC560C has the same edge length as the CDIL BC550C.
Yeah, I wonder about that, according to datasheets, their collector-base capacitances are identical. Their die geometries seem to be the same or very similar. Perhaps, they tried to make matching devices? If so, they succeeded. Or, may be, N-devices are only smaller when it comes to fets?
(https://www.richis-lab.de/images/Transistoren/82x01.jpg)
(https://www.richis-lab.de/images/Transistoren/71x01.jpg)
...
(https://www.richis-lab.de/images/Transistoren/71x07.jpg)
The structure is interesting. The left brown area is the p-doped emitter. The following green are is the n-doped base. In the next pictures we will see that this is the base-emitter-junction.
Before the base contact there is a small brown p-doped layer certainly above the base area. Why that? Looks like a pinch resistor to increase the base resistance. But why would they increase the base resistance? Perhaps the resistance is equalising the electric stress on the transistor area? :-//
(https://www.richis-lab.de/images/Transistoren/71x08.jpg)
That´s the base emitter junction! 8)
(13,5V / 20mA)
...
It seems a theoretical possibility that uniformly distributed base resistance reduces formation of current hot spots and improves SOA.
A local increase in current increases voltage loss across Rb and reduces Vbe.
(https://www.richis-lab.de/images/Transistoren/91x01.jpg)
That´s a dead Motorola MJ802 (90V/30A/2MHz/200W).
(https://www.richis-lab.de/images/Transistoren/91x02.jpg)
(https://www.richis-lab.de/images/Transistoren/91x03.jpg)
A big die and a big heatspreader as we would expect.
They used different bondwire diameters for emitter and base.
And yes, it´s dead. ;D
(https://www.richis-lab.de/images/Transistoren/91x04.jpg)
We can spot four dots due to the testing in the production line.
(https://www.richis-lab.de/images/Transistoren/91x05.jpg)
It´s a MESA
transistor with a trench at the edges.
(https://www.richis-lab.de/images/Transistoren/91x06.jpg)
(https://www.richis-lab.de/images/Transistoren/91x07.jpg)
Now that looks bad. There is a lot of molten metal.
(https://www.richis-lab.de/images/Transistoren/91x08.jpg)
After removing the bondwire we can see the whole mess. In the upper area there is a discoloration over the pn-junction like we have seen it in the 2N2222 I killed (https://www.richis-lab.de/Bipolar04.htm (https://www.richis-lab.de/Bipolar04.htm)). The first failure must have occurred in this area where most of the metal is molten. Some of the metal accumulated right of the bondwire. The die is cracked. :o
(https://www.richis-lab.de/images/Transistoren/91x09.jpg)
(https://www.richis-lab.de/images/Transistoren/91x11.jpg)
(https://www.richis-lab.de/images/Transistoren/91x10.jpg)
The cracks go all the way down to the heatspreader.
The molten trench is quite deep. Normally the metal layer isn´t that deep. I assume in this small area some of the silicon is molten. Silicon needs 1400°C to melt but with a lot of energy in this small area it should be possible.
(https://www.richis-lab.de/images/Transistoren/91x12.jpg)
I tried to remove the metal layer to see more details of the failed structures but the die fell apart...
https://www.richis-lab.de/Bipolar57.htm (https://www.richis-lab.de/Bipolar57.htm)
:-/O
And that was with the smaller SOT-227 modules. The bigger bus-bar mounted kind tend to launch shrapnel when this happens. Installing guards is recommended, or operating with the equipment closed up. :)
This is why most larger IGBT modules are gel filled. The gel absorbs the energy quite efficiently and doesn't create shrapnels.
This is why most larger IGBT modules are gel filled. The gel absorbs the energy quite efficiently and doesn't create shrapnels.
Yeah, that helps -- the difference is whether enough energy was released to blast the cover off it (and I suppose, launch bits of goo). Arc flash is serious business. :o
Basically the difference between a semiconductor fuse right at the device*, versus at the mains inlet (or any fuse type), is how much shrapnel is produced. ;D
*Not that you'd do this on a switching inverter, but upstream of minimal value bypass caps, perhaps.
Tim
Yeah, that helps -- the difference is whether enough energy was released to blast the cover off it (and I suppose, launch bits of goo).
That´s a dead Motorola MJ802 (90V/30A/2MHz/200W).
None of these had semiconductor fuses installed, just the usual line fuses, and the usual DC link capacitors ...
I've seen the aftermath of bigger booms than these, but the covers weren't blasted off, and the goo stays in place. Plasma and / or arcs can escape and leave a lot of blackened stuff, but no flying parts around from these modules. I've seen other stuff (like AC input bridge rectifiers, electrolytic DC link capacitors) explode more violently.
Yes, this stuff gets blown intentionally as part of product safety testing.
[...]
Most transistors use toroidal lines. Here we have squares with different size. :-//
[...]
[...]
Most transistors use toroidal lines. Here we have squares with different size. :-//
[...]
Perhaps a French design? French engineers often like to do things different... (think Citroën)
Interesting, I didn't know that Darlingtons were implemented on a single chip like that... makes sense, obviously.
Interesting, I didn't know that Darlingtons were implemented on a single chip like that... makes sense, obviously.Old story, Sidney Darlington had a patent on that.
8. A signal translating device comprising a body of semi-conductive material having therein a first zone of one conductivity type, a pair of spaced zones of the opposite conductivity type contiguous with said first zone and a pair of zones of said one type each contiguous with a respective one of said first pair of zones, and remote from the other, means electrically connectin one of said first pair of zones to the one of said second pair of zones remote therefrom, and individual electrical connections to said first zone, the other of said first pair of zones and the other of said second pair of zones.
Interesting, I didn't know that Darlingtons were implemented on a single chip like that... makes sense, obviously.Old story, Sidney Darlington had a patent on that.Quote8. A signal translating device comprising a body of semi-conductive material having therein a first zone of one conductivity type, a pair of spaced zones of the opposite conductivity type contiguous with said first zone and a pair of zones of said one type each contiguous with a respective one of said first pair of zones, and remote from the other, means electrically connectin one of said first pair of zones to the one of said second pair of zones remote therefrom, and individual electrical connections to said first zone, the other of said first pair of zones and the other of said second pair of zones.
Good observation. I wonder if that base-to-base resistor could make this device exempt from Darlington's patent :-DD
Perhaps not, because it had other, more general claims.
BTW, US2663806 was granted in 1953 so it should have expired long, long ago and not be a concern in any remotely modern design.
I see metal layer fully covers R1. Why it doesn't short it?
A long time ago SCS were used in a lot of special applications like counters and ring memory but even today there is an interesting applications shown in the datasheet of the TISP83121 which is still available. With such a protection circuit you don´t deflect overvoltages (and undervoltages) into your supply like with clamping diodes but to ground. That is often a lot less problematic.
If you did need an SCS now, a pair of bipolar transistors could be used just as with an SCR.
Long ago we used 3 SCRs with core die in planes X, Y and Z as high energy detectors. These were used to "crowbar" the main power supplies when a nuclear event was detected. The idea was too quickly discharge the main energy sources in the PS before they had time to damage the main electronics in the system. During an event all the semiconductor PN junctions become forward biased and can damage the chip, this also required a minimum decoupling capacitor to limit the available energy source for the forward biased junctions. We also used large die transistors like 2N3055 as detectors.
We had an on-site Flash Xray lab to emulate a nuclear event to help with these developments.
BTW great images as usual :-+
(https://www.richis-lab.de/images/transistoren/95x01.jpg)
...
We had an on-site Flash Xray lab to emulate a nuclear event to help with these developments.
I´m sure that was an interesting lab! ;D
Thanks!
Keep those great semiconductor images coming, these are fascinating to view, especially for those of us that worked in this field :-+
I still have quite a lot semiconductors: big and small, old and new, complex and simple, known by everyone and very special... 8)
I still have quite a lot semiconductors: big and small, old and new, complex and simple, known by everyone and very special... 8)
I vote for FRED/FERD diodes :)
(https://www.richis-lab.de/images/transistoren/97x14.jpg)
That´s interesting: The light in the driver transistor is quite wide.
You can see it´s no focus problem because in the corners there are sharp structures.
(https://www.richis-lab.de/images/transistoren/97x14.jpg)
That´s interesting: The light in the driver transistor is quite wide.
You can see it´s no focus problem because in the corners there are sharp structures.
Why does the light at the 4 corners located inside of the enclosed area (the 4 corners right near the bonding wire) appears as thick as in the straight lines, while in the rest of the corners the light appears to be thinner?
Just wondering, do beasts for 100 Amps and more still have the same structure just huge, or are many small in parallel?
(https://www.richis-lab.de/images/Transistoren/79x03.jpg)
I always found something oddly satisfying about those old power transistor packages.
... and the Fairchild transistors with circular ceramic base with black epoxy on top. When I was a kid repairing guitar amps, Peavy used these Fairchild transistors.
Think Fairchild also used this package with some RTL logic.
(https://www.richis-lab.de/images/transistoren/a10x05.jpg)Do I understand it right. Those tiny electrodes connected to bond pads are D/S and the die itself is Gate?
Well, Zeptobar´s pictures are better...
There is a very thin gate grid surrounding the drain and source areas contacted by the metal layer.
There are three round areas without contacts and gate grid. Would be interesting what these round areas do. Do they lower the current density? Do they optimize the manufacturing? Some weird noise reduction? :-//
:-/O
(https://www.richis-lab.de/images/transistoren/a10x05.jpg)Do I understand it right. Those tiny electrodes connected to bond pads are D/S and the die itself is Gate?
Well, Zeptobar´s pictures are better...
There is a very thin gate grid surrounding the drain and source areas contacted by the metal layer.
There are three round areas without contacts and gate grid. Would be interesting what these round areas do. Do they lower the current density? Do they optimize the manufacturing? Some weird noise reduction? :-//
:-/O
The 2AW was one of the cheapest BF862 I found on Ebay (0,5€ while the 2Ap was 2,74€).I'm shocked :-DD
The surface of the package looks a little strange like it was sanded.
Do I understand it right. Those tiny electrodes connected to bond pads are D/S and the die itself is Gate?Right. Basic structure shown below (Siliconix, "Designing with Field Effect Transistors"). In BF862 the gate is a grid rather than a single line or multiple lines. It seems that the substrate is typically (always?) connected to the top gate and also participates in pinching the channel.
I wonder if those "holes" in BF862 could be connections between the top grid and the substrate.
It absolutely makes sense.
Thanks! :)
Would be a good solution for a RMS detector. :-+
The more modern version is the LT1088:
https://richis-lab.de/LT1088.htm (https://richis-lab.de/LT1088.htm)
(https://www.richis-lab.de/images/transistoren/a14x03.jpg)Oh, just imagine that poor workers soldering wires to this :phew:
The wires were soldered to the connector pins... ...but that tin ball doesn´t look good. :o
(https://www.richis-lab.de/images/transistoren/a18x06.jpg)
(https://www.richis-lab.de/images/transistoren/a18x07.jpg)
It reminds of the rod and crankshaft from the Steam logo:
Dual transistors like that were very common as differential pairs until the 741 operational amplifier became popular and economical.
Dual transistors like that were very common as differential pairs until the 741 operational amplifier became popular and economical.
I should have mentioned that. The binned and thermally coupled transistors work well in differential stages.
A pair of P-channel MOSFETs.
Designed and manufactured by MNIPI.
This pair of transistors was used in electrometers and electrometric amplifiers developed by MNIPI.
An example of devices where these transistors were used are U5-11, V7E-42.
Of interesting, passport gate leakage current value does not exceed 1fA. The measured actual value of the gate leakage current is approximately 50aA.
What is 3n mosfet?
The substrate is connected to another emitter, rather.
A likely application for this arrangement (also CA3086, LM3086 etc.) is the substrate-emitter transistor serves as current sink, the common-emitter pair is the diff pair on top of it, and the other two can be used for whatever, maybe cascode or follower at the diff amp collectors. (Another monolithic pair would be needed for PNP current mirrors, if using that type of design. Would be handy if they put PNPs into the same package, but alas.) Gain depends on bias (no external resistance between emitters), so this is useful for mixers and variable-gain (OTA) circuits as well as plain old amplifiers.
Have you ever heard of the company "Solid State"? Try to search for information about a company named "Solid State"! ;D
...
The 2N3752 is a power transistor in a TO-111 package that can be screwed into a heat sink and dissipate up to 30W of power at 100°C. It can isolate 80V and conduct up to 1A. Cutoff frequency is 50MHz. The transistor is isolated from the package, which has four connections for this reason.
The 2N6485 die shot reminded me of this one of INT591A from Zeptobars -
https://zeptobars.com/en/read/1NT591A-double-bjt-transistor-array-differential-amplifier (https://zeptobars.com/en/read/1NT591A-double-bjt-transistor-array-differential-amplifier)
GD170, a PNP-Ge-Transistor built in the Halbleiterwerk Frankfurt Oder. 30V/3A/180kHz/5,3W@25°C and a hfe between 18 and 90. ;D
In fact nothing very special.
It could have especially low saturation. :)
Germanium transistors do have lower Vce(sat) than silicon transistors.
...
Focus stacking taken to new heights! :D
With a narrow aperture you get problems at high magnifications due to diffraction. So you better work with wide apertures, a lot of pictures and focus stacking.
With a narrow aperture you get problems at high magnifications due to diffraction. So you better work with wide apertures, a lot of pictures and focus stacking.
But that would be cheating.
With a narrow aperture you get problems at high magnifications due to diffraction. So you better work with wide apertures, a lot of pictures and focus stacking.
But that would be cheating.
Motorola 2N3614, 35V, 7A continues, 15A peak, 77W @25°C case temperature.
There are two metal sheets contacting base and emitter which were initially connected.
(https://www.richis-lab.de/images/transistoren/a35x06.jpg)
(https://www.richis-lab.de/images/transistoren/a35x07.jpg)
There is a n-doped germanium slice acting as base region. The p-doped emitter and collector areas are alloyed into the two sides of the base slice.
(https://www.richis-lab.de/images/transistoren/a35x04.jpg)
(https://www.richis-lab.de/images/transistoren/a35x05.jpg)
There are two metal sheets. In the first place the base metal sheet is fixed by the two contact pins. The emitter metal sheet is fixed by the emitter pin and a hole in the base metal sheet. Taken together this construction assures that the emitter contact is in the middle of the germanium slice. The base metal sheet is cut at the hole where there is only one metal sheet and the cross section is lowest.
https://www.richis-lab.de/Bipolar86.htm (https://www.richis-lab.de/Bipolar86.htm)
:-/O
With a narrow aperture you get problems at high magnifications due to diffraction. So you better work with wide apertures, a lot of pictures and focus stacking.
I have added 3TB to my NAS. ;D
The ГT906 (GT906) is a germanium power transistor that can block up to 75V. ... The second line could be a date code (1981).
I don't know for the OC45 but it seems there was a lot handcraft back in the days:No wonder it used to cost a fortune
https://www.thevalvepage.com/trans/manufac/manufac1.htm (https://www.thevalvepage.com/trans/manufac/manufac1.htm)
I don't know for the OC45 but it seems there was a lot handcraft back in the days:No wonder it used to cost a fortune
https://www.thevalvepage.com/trans/manufac/manufac1.htm (https://www.thevalvepage.com/trans/manufac/manufac1.htm)
The thinnest base wafers are those produced for the OC44 and OC45.These wafers are only 100 µm thick - a tenth of a millimetre. At this still quite early stage in manufacture, the dice are worth more than their weight in gold.
That layout equalizes the emitter resistance to the far points of the emitter.
Just found this.
(Attachment Link)
I would like to see what a Zetex "super e-line" transistor looks like. Other manufacturers now make them as low saturation voltage and high current gain parts.
ZTX689B (https://www.diodes.com/assets/Datasheets/ZTX689B.pdf) and ZTX851 are popular not as a medium power transistor, but for low noise audio preamplifiers and they have a low rbb'. The die would be very interesting to see.
ZTX689B (https://www.diodes.com/assets/Datasheets/ZTX689B.pdf) and ZTX851 are popular not as a medium power transistor, but for low noise audio preamplifiers and they have a low rbb'. The die would be very interesting to see.
(https://www.richis-lab.de/images/transistoren/a39x01.jpg)
Sescosem 391HT2 - You don´t find any information about this transistors. It is said to be similar to the BDY25: 140V, 6A, 88W, 10MHz
[...]
https://www.richis-lab.de/Bipolar90.htm (https://www.richis-lab.de/Bipolar90.htm)
I understand about passivation but would that be passivation of the metal parts? I guess there is not much point in passivating germanium or silicon?
Anyway, unfortunately i could not retrieve my transistor of the bottom of the garbage can, my girlfriend tossed the contents of the bag-less vacuum cleaner on top of it and i am not looking forward to wonder through a few pounds of vacuum cleaner dust and crap, sorry :--
It wasn't all that exciting also just a piece of germanium with 2 contact points on it.
(https://www.richis-lab.de/images/transistoren/a50x01.jpg)
https://www.richis-lab.de/Bipolar99.htm (https://www.richis-lab.de/Bipolar99.htm)
As always, nice image, and amazed at how many of these you produce :-+
I would like to see what a Zetex "super e-line" transistor looks like. Other manufacturers now make them as low saturation voltage and high current gain parts.
I would like to see what a Zetex "super e-line" transistor looks like. Other manufacturers now make them as low saturation voltage and high current gain parts.
Had a spare ZTX851, unfortunately the die cracked while decapping but most of it is still intact.
Die is aprox 1780x1780µm and has a perforated emitter design - layout looks very similar to the 2STD1665 (https://zeptobars.com/en/read/ST-2STD1665-NPN-BJT).
Hm, I forget if Early effect is especially worse on these.
Is the base layer unusually thin (and fT and hFE maintained by the fine interdigitation)?
https://www.youtube.com/watch?v=AGXzE6qgipw (https://www.youtube.com/watch?v=AGXzE6qgipw)
No further information can be found. I assume the type designation consists just of the characters A479A. 67 could refer to the year 1967. :-// The component tester says it is a silicon NPN transistor.The marking of the transistor is non-standard, which means an experimental sample. The first letter "А" (Cyrillic) denotes the plant - Pulsar. "А479А" (Cyrillic) is most likely a prototype of the transitor 2Т319А (Cyrillic).
No further information can be found. I assume the type designation consists just of the characters A479A. 67 could refer to the year 1967. :-// The component tester says it is a silicon NPN transistor.The marking of the transistor is non-standard, which means an experimental sample. The first letter "А" (Cyrillic) denotes the plant - Pulsar. "А479А" (Cyrillic) is most likely a prototype of the transitor 2Т319А (Cyrillic).
Here is information about it: http://www.155la3.ru/2tp319.htm (http://www.155la3.ru/2tp319.htm)
How do you get from A479A to 2T319A?Google gives out some information from a closed forum. Plus the production years are similar.
If you ever get the opportunity to drop it in front of an XRD analyzer, that should tell something. ;D
Aluminum in particular cannot be bonded to gold (due to formation of an especially brittle intermetallic), but a more compatible or more resistant metal could be used. I'm not sure what, offhand; not nickel either, as that has the same problem; tin is well-behaved with both, but that's just a soldered joint and may not be strong enough; iron may be okay (intermetallic is still formed, but a harder one, in a thinner interface layer I think, and importantly, it won't tend to degrade over time as for more mobile ions (lower melting substances)); or perhaps harder/refractory metals are used -- I don't know about say chromium or molybdenum, but maybe aluminum just acts like solder on them, not reacting enough to matter?
Noopy, I have some very old chips from ECI (E-systems) in the USA. They are ceramic carriers with gold (plated?) terminals. The chip seems to be coated in some PMMA and/or glass encapsulation. Would you be interested in photographing some of them?
(https://www.richis-lab.de/images/transistoren/a59x01.jpg)
I ordered a IRF3708 at Reichelt and in my view that is a counterfeit part. The IRF3708 is a common power MOSFET which is now obsolete and because of that it´s hard to get genuine parts.
...
On a CMOS sensor with the IR filter removed there is just enough IR sensitivity to see the IR light emitted by PN junctions in forward bias.Interesting possibility.
The die of a 2N3439 transistor in visible light, as well as photos showing the light emissions under Emitter-Base breakdown, and the IR light emitted when the base junction is forward biased. On a CMOS sensor with the IR filter removed there is just enough IR sensitivity to see the IR light emitted by PN junctions in forward bias.
Very interesting stuff. So does this mean that effectively, the 'resistance' in a semiconductor 'happens' at the junction - and that's why it gets hot there? I.e. there is no way around Ohm's law, even for a semiconductor (there is equivalent "resistance" at the junction, and putting a current through it will make it hot).
Just a small Philips BC548C...
Vce is a little lower than for the BC547C (https://www.richis-lab.de/Bipolar44.htm (https://www.richis-lab.de/Bipolar44.htm)).
Just a small Philips BC548C...
Vce is a little lower than for the BC547C (https://www.richis-lab.de/Bipolar44.htm (https://www.richis-lab.de/Bipolar44.htm)).
I always thought these were the same production but graded for Vce. My own tests with a curve tracer showed that the lower voltage versions often met the higher voltage breakdown specifications.
How about a transistor-die-picture-topic? :)
I have collected some here:
https://www.richis-lab.de/Transistoren.htm (https://www.richis-lab.de/Transistoren.htm)
And I just have to show you this one:
(https://www.richis-lab.de/images/Transistoren/02x09.gif)
You see the breakdown of the KD501-base-emitter-junction with increasing current. 8) ;D
(https://www.richis-lab.de/images/Transistoren/02x07.jpg)
Does anybody know why Tesla integrated this step at the edge of the die?
Is there a right way to open TO 220 to see the die? I want to compare both Chinese TDA2050 and LM1875.
(https://www.richis-lab.de/images/Transistoren/50x03.jpg)
...
(https://www.richis-lab.de/images/Transistoren/50x09.jpg)
https://www.richis-lab.de/Bipolar65.htm (https://www.richis-lab.de/Bipolar65.htm)
So... Which 2N2222 is the best? :)Well, the Motorola one is a mutant.
So... Which 2N2222 is the best? :)
2N4401 ;D
So... Which 2N2222 is the best? :)
Well, the Motorola one is a mutant.
The 'Mutant' stood for 'Marvel's X-Men' from their comics / movie cinematic universe.2N4401 ;D
I concur.So... Which 2N2222 is the best? :)
Well, the Motorola one is a mutant.
The Motorola one shows the structure of the original 2N2222 which gave it its performance. The others rely on better processes.
So... Which 2N2222 is the best? :)
Well, the Motorola one is a mutant.
The Motorola one shows the structure of the original 2N2222 which gave it its performance. The others rely on better processes.
The 'Mutant' stood for 'Marvel's X-Men' from their comics / movie cinematic universe.
A reversed BE can be used as a noise generator. I wonder how much noise correlation would be if each transistor in such a pair would be used as an independent noise generator. ???
these die photos are phenomenal! nice work Noopy!
(https://www.richis-lab.de/images/transistoren/a85x01.jpg)Looking at the circuit, diode Q6 is actually there for voltage drop without additional current gain. If you just had Q9 without the diode in series with its base, the emitter follower formed by Q5 that comes off the output of the error amplifier may not be able to turn Q9 off; the output device of the error amplifier (Q8) goes into saturation (~100mV depending on process) and you go up 0.7V from the Vbe of Q5 and will have ~0.8V at the base of Q9 no matter how hard you want to turn him off. If you made Q6/Q9 a Darlington, you could potentially have too much current gain and loop instability.
The TDE1737 is a relay and lamp driver like the TDE1647. While the TDE1647 contains a highside output stage, the TDE1737 has a lowside output stage. The reverse voltage is up to 50V, the current may rise up to 1A.
(https://www.richis-lab.de/images/transistoren/a85x02.jpg)
Pretty similar to the TDE1647.
(https://www.richis-lab.de/images/transistoren/a85x03.jpg)
The differential amplifier at the input (yellow) is similar to the one in the TDE1647. However, the common current source is missing here. The current through the branches is adjusted just via the transistors Q3/Q4. The bias setting (blue) and reference current sink (cyan) look the same like in the TDE1647.
The output of the differential amplifier passes two amplifier stages (pink). The second amplifier stage is reminiscent of a Darlington transistor. However, Q6 is connected as a diode. Either one wanted to provide the possibility to build up a Darlington transistor or Q6 ensures that Q9 switches off more slowly and the output stage switches on with a minimal delay. In addition, it could of course be a remnant of the TDE1647 circuit, which is quite similar to the TDE1737.
As long as the transistor Q9 is inactive, the current of the current source Q15 controls the output stage, which consists of a Darlington transistor (red). As overcurrent protection, transistor Q24 (green) diverts the base current of the output stage if too much voltage drops across the external shunt.
The overtemperature protection of the TDE1737 (grey) is based on the reference voltage of the reference current sink (cyan). If the temperature rises, the base-emitter voltage of Q18 drops and a current flows through Q18, Q21 and Q17. The current mirror Q17 then controls Q16 and diverts the base current of the output stage through it.The way I look at this is operating voltage range. Diode D2 is a Zener, so if you look at the currents from Q15 you will see one goes to drive the output transistor, one goes to Q22 base and Q23 collector, and the last goes through D2 to the base of Q23. Consider the voltage across Vce(sat) of Q15 in series with the D2 Zener voltage (likely 5.5-6V range) and Vbe of Q23: until this voltage is high enough, D2 will not conduct and Q22 will be on, holding off the output. Now, when you change to the two diodes just in series with Q15, it only takes the four diode drops plus the Vce(sat) of Q15 to run, so the part will turn on at a lower input voltage.
The remaining elements (purple) appear to perform a similar function to the ground loss protection circuit in the TDE1647. Ground loss is less critical in a lowside driver. However, the circuit could protect the TDE1737 against supply voltages that are too low to guarantee a proper functionality. Only when the voltage between Vcc and GND is high enough for D2 to conduct, Q23 diverts current from current source Q15, thus deactivating transistor Q22, which otherwise, like transistor Q16, keeps the output stage inactive.
(https://www.richis-lab.de/images/transistoren/a85x04.jpg)
(https://www.richis-lab.de/images/transistoren/a85x05.jpg)
The dimensions of the die are 2,1mm x 1,4mm. On the lower edge, a relatively large logo refers to ST Microelectronics.
(https://www.richis-lab.de/images/transistoren/a85x06.jpg)
On the upper edge there is a year which is difficult to identify. It looks like 1985.
(https://www.richis-lab.de/images/transistoren/a85x08.jpg)
In the lower left corner, the characters 1737.C are shown in the metal layer. The C could stand for a third revision of the design.
(https://www.richis-lab.de/images/transistoren/a85x07.jpg)
(https://www.richis-lab.de/images/transistoren/a85x13.jpg)
In the upper left corner the mask revisions are shown. Two masks are difficult to see, but as with the TDE1647, there are seven masks in total.
(https://www.richis-lab.de/images/transistoren/a85x09.jpg)
X057 could be an internal project designation.
(https://www.richis-lab.de/images/transistoren/a85x10.jpg)
A closer analysis of the dies reveals some minor differences to the schematic. It is also noticeable that the design is very similar to the TDE1647 in many places.
(https://www.richis-lab.de/images/transistoren/a85x11.jpg)
The most noticeable thing is that the purple circuit is missing completely. Instead, two diodes have been integrated into the collector path of the current source Q15. The additional voltage drop probably ensures that the output stage can only be controlled when the supply voltage reaches a value at which the rest of the circuit can do it´s job properly.
(https://www.richis-lab.de/images/transistoren/a85x12.jpg)
On the upper edge there is a bondpad that cannot be assigned to any function. The current source Q14 has an additional emitter whose constant current is led to a small circuit below the bondpad. There is the NPN transistor Qa and the resistor R with three taps.Another possibility is that it is a function that is included in a different product where the pin is used as well? As suggested, with a different metal layer, this could be added into the circuit. Maybe it was a way to add hysteresis to the part? Considering it is right next to the non-inverting input (right where you would want to introduce hysteresis), that could be a useful bit.
(https://www.richis-lab.de/images/transistoren/a85x14.jpg)
It remains unclear what function the additional circuit has. Perhaps it enables to determine the manufacturing quality. Perhaps it is a hold-off that only has a meaningful function with a different metal layer.
https://www.richis-lab.de/BipolarA25.htm (https://www.richis-lab.de/BipolarA25.htm)
:-/O
Think about me peeling this tiny thing out of a black block of burned epoxy.
After that there is some cleaning and putting it in front of the camera in the right direction. ;D
(https://www.richis-lab.de/images/transistoren/a94x01.jpg)
Huh, and fT isn't higher? Also their fT curve doesn't meet the minimum? :-DD
...Hmm interesting, few others have a curve but one that does (onsemi) also shows just under 300MHz ...at Vce = 1V?! Weird, go figure. :D
The edge length of the die is 0,27mm. This makes the die the same size as most simple small signal transistors. In such transistors, simple designs are usually found in which an emitter area is integrated into a base area. Here, on the other hand, two base areas have been structured, each containing three narrow emitter areas.
The edge length of the die is 0,27mm. This makes the die the same size as most simple small signal transistors. In such transistors, simple designs are usually found in which an emitter area is integrated into a base area. Here, on the other hand, two base areas have been structured, each containing three narrow emitter areas.
As always, thank you for posting these. I'm trying to figure out exactly what each part actually is. Looking at the second picture, what exactly do the two green squares represent? Is that the P-type base material?
What makes me very suspicious is that their numbers 1:1 match with TI datasheet: https://datasheet.lcsc.com/lcsc/2005151136_Texas-Instruments-NE5532ADR_C529290.pdf (https://datasheet.lcsc.com/lcsc/2005151136_Texas-Instruments-NE5532ADR_C529290.pdf) .But they don't match the TI datasheet exactly. There are differences such as slew rate and supply current (which also seems specified per channel rather than per package) and errors - their 30nA typical Ibias spec looks unilkely, for example.
An interesting detail: Chinese DS provide phase and gain plots which are not present in others' datasheets. But of poor quality, like a copy of a bad copy. Very suspicious.
Anyway, as far as second-sourcing, the easiest way to sell your parts is to specify them as the originals were specified. Copy the min/max parameters verbatim, and work to match them as well as is possible (or reasonable). Maybe lying about some parameters is "okay" (strictly in the business sense, not engineering!), but preferably the typicals are characterized and listed earnestly, and maybe they vary a bit from the originals but for typ that's okay.
Saw one example the other day,
http://file.3peakic.com.cn:8080/product/Datasheet_LM2903A-LM2901A.pdf (http://file.3peakic.com.cn:8080/product/Datasheet_LM2903A-LM2901A.pdf)
Almost certainly a 100% CMOS version, NOT a drop-in equivalent; but looks reasonably useful, and compatible in a lot of applications. Naming it identically is disingenuous and I'm sure will catch some off guard.
How is the SCT2450KE bonded to its package? I didn't see a leadframe.
Very cool! AFAIK, SiC are currently not Superjunction, just traditional planar stripe VDMOS or whatever. The top grid pattern presumably is source connections (vias) dipping down to the die surface, through a grid (square mesh?) of gate connections. Analogous to ye olde HEXFETs, as far as the top interconnect goes. But with a square grid, and I assume a vertical trench structure below. It's not clear if the FET cells would also be on a square grid, or are just stripes the whole width.
The size comparison is particularly apt, as the Si chips are also traditional planar technology -- which puts them at particular disadvantage at the higher voltage ratings. For that technology, specific Rds(on) scaled something like Vds^2.2.
Thanks to the high breakdown field strength, comparable geometry (e.g. junction depletion widths) would give more like a 100V rating in Si, but more like 1000V in SiC. That's basically a IRF520!
Oh! I don't suppose that's still working to some extent? Maybe not after the acid dip, but, I wonder if the body diode glows in forward bias? >:D
I hope you weren't using a bench supply (CC/CV with big capacitor on the output) for that test...were you? :horse:
Tim
I wonder how they implemented the diode. P-region buried under the emitter contact, perhaps? There's a faint outline under the emitter metal, but that could just be connections.
I wonder if a parasitic SCR (PNPN) breakdown mode could be activated, say with large dV/dt, or forward or reverse avalanche...
BUX22 rated 250V 40A 350W hFE 20-60, fT 10MHz "silicon multiepitaxial planar NPN".
The collector current may be up to 150mA. At a case temperature of 100°C, the TO-39 case can dissipate up to 6W. The DC gain is specified as 60 (100V/50mA). The cut-off frequency is 145MHz.
Hmm, not so much heavily doped, given the modest breakdown voltages; gold doping likely though?
Could well be higher doping in the connection layers (by extension, substrate -- being the collector connecting layer -- if epitaxial type), forcing recombination near the junctions and reducing switching resistance.
(...Ah yes, "double diffused epitaxial", so that could be. That is, n+ substrate; n- epitaxy for collector drift region; p (base) then n+ (emitter) diffusions; and, maybe p+ base contact diffusion?, but would that be triple, or can it be done at the same time as emitter? Not sure. ...Nah, don't think there's a base contact layer. That checks out then.)
Also I guess an extra "emitter" ring around everything, for guard ring? I... forget exactly how those work, but that looks like what they've done anyway.
Unglaublich saubere Fertigung. für das Alter. Haben wir Restströme von dem Ding ?
Thanks, I will try to get some parts that are still in their can. Opening normally kills leakage current due to moisture and other gases.
Where did you get your parts from ?
I was expecting a much bigger R from such a thin wire.It burns very easily turning into plasma. But the plasma channel has low resistance.
Interesting story, beautiful pics, thanks! :-+When you're talking short distances, resistances turn out to be fairly low value. A gold bond wire that is 1mil diameter ends up at about 4.5 milliohm. What is the big deal on it isn't so much the resistance, but instead the fusing current. That's dependent on a number of variables from mass/cross sectional area to power dissipation and time. Sitting in open air like those will also lower the fusing current (over-molded bond wires can handle higher currents).
Just out of curiosity, estimating R of a tin whisker:
R = \$\rho\$*L/S
\$\rho_{\mathrm{Sn}}\$ = 10.9E-8\$\Omega\$*m
L = 0.1mm
d = 5um
R = 10.9E-8 * 0.1E-3 / (3.14 * 25E-12 / 4) = 1.09/(3.14*1.25) = roughly 1/4 = 0.25\$\Omega\$ ???
I was expecting a much bigger R from such a thin wire.
Here you can see the germanium power transistor OC864A, which was developed in the Funkwerk Erfurt but never went into series production. Only a few sample exist in the Thuringian Museum of Electrical Engineering (https://www.elektromuseum.de (https://www.elektromuseum.de)). The following background information also comes from this museum.
In 1959, the VEB Funkwerk Erfurt (FWE) ceased production of transmitter tubes. The capacities freed up were used to start developing germanium power transistors in the so-called Zentrallabor für Empfängerröhren (ZLE). An important basis for this work was a Soviet documentation. In addition to the necessary dimensions of the germanium crystal, it described how to etch the material, which alloy materials to use and which geometries to aim for. Alloying was carried out in a graphite mould under vacuum. The design of the alloying furnace was also taken from the Soviet documentation. However, the optimum temperatures and times for the alloying process were missing and had to be determined from tests. At the end of 1962, after the development had been completed, the results were transferred to the Halbleiterwerk Frankfurt Oder (HFO). [...]
"An important basis for this work was a Soviet documentation. "
High purity germanium crystal growing tech was a big deal back then. I'm curious who really had it or was it simply stolen IP?
Bell Labs 1953 William Pfann (https://en.wikipedia.org/wiki/William_Gardner_Pfann) developed "zone refining" and I saw Japan using it to make transistors in the early 1950's.
As in the Siemens 2N3055 from 1975 (https://www.richis-lab.de/2N3055_01.htm (https://www.richis-lab.de/2N3055_01.htm)), there is a white powder in the package, which presumably is a drying agent.
Are you sure you found the thermal paste in TO3 transistors?It is not a thermal paste.. It is an expressionistic vision of the late twentieth's century technology progress ("Controlled Currents in a Pot")..
Are you sure you found the thermal paste in TO3 transistors?