EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: OM222O on September 21, 2018, 01:55:11 am
-
Hello, I was trying to build a dummy load and it seems everyone is recommending to use a mosfet in it's linear region to control the current and therefore dissipate all the heat.
As far as I know, IGBT is basically a mosfet with a pass transistor on it's output. They have way higher voltage and current ratings as well as wider safe operation area, they supposedly react faster than mosfets and in most cases are cheaper than a suitably rated mosfet for a given power target.
Is there a reason that nobody is using them and instead they use mosfets? in which case would it make sense to create the same effect using a discrete mosfet and BJT?
are there any drawbacks to using IGBTs? they seem to be the superior choice. The only think I can think of, is not being able to go below ~0.7 volts which is not an issue for me as even then 1.5v alkaline batteries have a cut off voltage of 0.8 volts. I think they're also easier to cool compared to mosfet as they're designed and manufactured with that voltage drop in mind witch makes them even more suitable in a situation where they're only purpose is to transfer electrical energy to heat?
-
No linear SOA.
IGBTs are constructed expressly to have the most current density possible, damn anything else -- which means you put any voltage across them and they melt real damn quick!
You could, however, make a switching dummy load, where the input goes through an inductor, the IGBT chops the current through the inductor, and the flyback is dumped into a fixed dummy load, like a resistor or MOV or something. Or recycled into a source battery, say.
Tim
-
I can't quite understand what you mean ...
The SOA for mosfets isn't perfectly linear either?
and I'm not sure what you mean by:"you put any voltage across them and they melt real damn quick"
As I mentioned, they're rated for much much higher voltages than a mosfet. for example the "IKW15N120BH6XKSA1" is rated for 1200 volts! that seems insane compared to a few hundred volts maximum for a mosfet. I know they're meant for the most current density, which is exactly why I think they're a good candidate for a high power dummy load. I'm trying to make a dynamic electronic load, so the fixed method is not really good at all. I'm planning on using 4 of the chosen device (either IGBT or MOSFET), each one controlled via an op amp and a MCU) so I can spread the heat and check each one's temperature using NTCs to balance current and heat output between the 4. I'll most likely aim for 50 volts, 20 amps, 200 watts and spread the 4 into 2 groups of 2 (again for increased surface area and better cooling), each cooled by a CPU cooler, so each device needs to handle roughly 50 watts which is not that large.
I'm not planning as using this for a current source or anything, specifically testing PSUs and batteries, hence the MCU to allow for data logging. It would be great if you could explain a little more about what you wrote. thanks
-
Toshiba used to make IGBTs for linear service in audio power amplifiers but that was years ago and only up to a couple hundred volts. They even made complementary pairs!
IGBT safe operating area is not any better than that of MOSFETs. If they are not showing it in the datasheet, then it is because they are not intended for linear operation; it is not square or even flat. Even worse they use minority carriers like bipolar transistors which allows for a much smaller die (lower cost) for a given current and voltage, especially voltage, but this also results in a lower power rating because of the small die size.
The big advantage of IGBTs over MOSFETs is that at high voltages, MOSFET die size is proportional to the square of the voltage. IGBTs like bipolar transistors rely on minority carriers so they so not suffer from this limitation and can have a much smaller die size. Bipolar transistor are more economical though.
If you are building big 1000 volt electronic loads, then you do not need my help. You will require lots of devices in series and parallel to distribute the power and handle the worst case safe operating area no matter what power pass element you use. Those big linear MOSFETs from IXYS start to look like a deal if they cut the number of pass devices down.
-
"No linear SOA" as in, no DC (linear operating mode) SOA curve at all, or if there is one (I forget if I've ever seen a DC curve on an IGBT), it is tiny, a small fraction of the total power rating.
MOSFETs are available up to 4500V -- with pretty reasonable ratings at that, too* -- I'm not sure where you're finding "a few hundred volts maximum". :-//
*I used to observe that, heater power aside, vacuum tubes still held their own in this one domain: at high voltages and modest to very high powers. For example, a 6LQ6 horizontal output tube might have a saturated plate "resistance" of 70 ohms, and a peak voltage rating of about 7kV (strictly under cutoff conditions, mind). The heater and screen power requirements make overall efficiency rather worse of course (about 20W between them, plus a plate dissipation rating of 30W), nevermind the requirement for two extra supplies (heater and screen), plus a fairly large grid voltage (about -200V is recommended in peak cutoff).
Over the last 5 years or so, MOSFETs have been released with ratings around 2500V 10A and 4500V 2A (30 ohm?), completely removing any claim tubes might've still held in this domain. :)
(Vacuum tubes for high frequency (klystrons, TWTs, etc.), high power (some industrial and radio finals; magnetrons, etc.), and various physics and research applications, still dominate those fields. Of those, TWTs are probably the next to fall: they've been flying on satellites for literally as long as satellites have been flown in space, and are extremely well understood, reliable and still perform quite well despite their old fashioned nature. However, microwave transistors -- Si LDMOS, GaAsFET, PHEMTs and GaN FETs -- have been in commercial use for many years now, and it's my understanding that some are finally getting rad-hard and space-ready approval..?)
Ahh... I digress. Fun stuff.
Tim
-
This IGBT is fairly cheap and seems fit for the job. It has a SOA chart unlike what previously has been said, and given the numbers I provided, it seems like a fairly decent choice.
https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?qs=sGAEpiMZZMv4z0HnGdrLjmAnkg%2f6XuvzWCGzpW%2fDvWZ3y7weEXwbsA%3d%3d (https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?qs=sGAEpiMZZMv4z0HnGdrLjmAnkg%2f6XuvzWCGzpW%2fDvWZ3y7weEXwbsA%3d%3d)
(https://i.imgur.com/iunNkOC.jpg)
Can you please confirm I'm not making a mistake in reading the datasheet? It has way better SOA for similarly priced mosfets.
(Please keep in mind I'm limiting the power of each part to 50 watts, so @ 50 volts 1A would be the current limit and as you decrease the voltage, the current can go up accordingly, so my intended area of operation falls way below the provided SOA)
-
Well, if it's in the SOA, there you go...
I'd test it very carefully to be sure, though.
Tim
-
Don't forget temperature derating. That SOA curve is with a nearly impossible to hit 25C Tcase. What's your heat sink you plan on using?
-
I'll be using cpu coolers rated for 100 watts to cool 2 IGBTs each with a load of 50 watts under the heat sink. I'm expecting tempretures in the high 70s or mid 80s wich I believe still gives me a good enough margin. I'll be dissapating about 5 watts in the shunt resistor under max load which drops the actual IGBT load to 45 watts (so a total of 90). I'll be monitoring temperature of each individual IGBT with an NTC thermistor (10k THT mounted in the hole of the TO-247 packaging which gives accurate enough temperature readings when compensated for the package thermal resistance (which again, in this case is way less than a similarly rated FET). I still didn't get a proper answer tho: given the superior performance in every aspect when compared to a fet, why aren't IGBTs used in every single electronics load? other than SOA no other parameters were brought up?
-
Why not BJTs? :-//
-
there are two more aspects icould think of:
1. VCESAT vs. RDSON: if you want to go to really low load voltages (say, 2V and below), an IGBT isn't probably the best choice because of the always present VCESAT. also, datasheet VCESAT is dependent on the actual speed rating of the IGBT: slow IGBTs will reach down to 0.9 V, whereas fast IGBTs may have VCESAT beyond 2V. so, if you don't require too fast transient responses of your load, you would chose a slow IGBT to shave off some of the VCESAT.
2. MOSFET hot-spotting: although RDSON does have a positive tempco, which greatly helps in paralleled _switching_ applications, the VGSth actually has a negative tempco, which can be desastrous in _linear_ applications. a MOSFET itself consists of thousands of paralleled cells on a single MOSFET die. if the MOSFET runs in the VGSth region (as in your application), some of the cells in that sea of cells may get a little hotter than the surrounding ones, which in turn reduces VGSth in that area, which then increases the local drain current, which increases the local temp,... you get the picture. countermeasure is indeed heavy heatsinking, and ample over-dimensioning of the MOSFET.
idk, IGBTs may have the same effect, perhaps other forum members with more experience on IGBTs can comment on this.
hth
-
2. MOSFET hot-spotting: although RDSON does have a positive tempco, which greatly helps in paralleled _switching_ applications, the VGSth actually has a negative tempco, which can be desastrous in _linear_ applications. a MOSFET itself consists of thousands of paralleled cells on a single MOSFET die. if the MOSFET runs in the VGSth region (as in your application), some of the cells in that sea of cells may get a little hotter than the surrounding ones, which in turn reduces VGSth in that area, which then increases the local drain current, which increases the local temp,... you get the picture. countermeasure is indeed heavy heatsinking, and ample over-dimensioning of the MOSFET.
idk, IGBTs may have the same effect, perhaps other forum members with more experience on IGBTs can comment on this.
That's the mechanism, but it doesn't much matter for practical purposes: it's rolled into the SOA. At least, hopefully, if the manufacturer has done their due diligence...
Why not BJTs? :-//
Almost as bad as IGBTs.
BJTs for amplifiers are made to avoid hot-spotting and therefore 2nd breakdown. These aren't available much beyond 250V. Switching BJTs (rated 1500V or more) are optimized in the opposite way, giving rather small SOAs.
Tim
-
Why not BJTs? :-//
just my 0.02:
have a look at the datasheet of one of may favorites:
https://www.digchip.com/datasheets/parts/datasheet/456/BUT30-pdf.php (https://www.digchip.com/datasheets/parts/datasheet/456/BUT30-pdf.php)
this is a real unit in current handling capability and power dissipation, so it seems to be a good choice. however, BASE current requirement of several amps and the fact that it is absolutely not up to linear operation (look at the SOA curve: just 2 A IC at 20 V VCE: pathetic!) basically prohibits its use in electronic loads.
-
@: Tim: beat me to it! ;D
-
Why not BJTs? :-//
just my 0.02:
have a look at the datasheet of one of may favorites:
https://www.digchip.com/datasheets/parts/datasheet/456/BUT30-pdf.php (https://www.digchip.com/datasheets/parts/datasheet/456/BUT30-pdf.php)
this is a real unit in current handling capability and power dissipation, so it seems to be a good choice. however, BASE current requirement of several amps and the fact that it is absolutely not up to linear operation (look at the SOA curve: just 2 A IC at 20 V VCE: pathetic!) basically prohibits its use in electronic loads.
What about oldy BU208 and BU508 transistors? Linear 1500V TO3 package
-
dsheeesus, why aren't you just looking up the datasheet numbers? :-DD
-
dsheeesus, why aren't you just looking up the datasheet numbers? :-DD
:-//
-
This IGBT is fairly cheap and seems fit for the job. It has a SOA chart unlike what previously has been said, and given the numbers I provided, it seems like a fairly decent choice.
https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?
(https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?qs=sGAEpiMZZMv4z0HnGdrLjmAnkg%2f6XuvzWCGzpW%2fDvWZ3y7weEXwbsA%3d%3d)
That's interesting, most IGBT's don't promise anything for DC SOA.
Fairchild's trench stop technology based IGBT's seem to have DC SOA specification, added benefit of trench stop or just hopefull specification?
-
We once tried to use a real bad-ass IGBT that's been left over form another project in linear mode as a load, far below it's rated maximum dissipation (10kW). Sure enough, no DC SOA data was available. It died at a single-digit percentage of it's rated dissipation in switching operation. So I guess, unless there's proper data available, consider an IGBT as a switch and not as a resistor.
-
If you are building big 1000 volt electronic loads, then you do not need my help. You will require lots of devices in series and parallel to distribute the power and handle the worst case safe operating area no matter what power pass element you use. Those big linear MOSFETs from IXYS start to look like a deal if they cut the number of pass devices down.
At that high voltage, wouldn't tubes start making sense?
-
Nah, there are 2500V+ FETs on the market, cheaper than sweep tubes (the audiophiles and TV restorers keep their prices surprisingly high) and far easier to use. If nothing else, lower voltage parts can be cascoded.
Might be tempting for stupid high voltages (10kV+, single digit to fractional amperes?), but those transmitter tubes aren't any cheaper, and if you're doing quite that much power you're probably a power company that has a better way of dealing with that anyway (e.g., braking resistors?). :)
Tim
-
Using a vacuum tube might still be attractive at high voltage, as they tend to fail open - while semiconductors tend to fail short. However the minimum working voltage is rather limited.
...
What about oldy BU208 and BU508 transistors? Linear 1500V TO3 package
The BU208 / BU508 are not that good SOA wise. The right MOSFET types (usually higher voltage types even for just 30 V use) are still the better choice at higher voltages. The BJTs (especially audio types) might be a good choice up to about 50 V, maybe 100 V)
-
I wouldn't subscribe to that: If a vacuum tube is severely overloaded, usually the vacuum gets compromised (crack, seal breach, outgassing of internal components) which will eventually lead to a discharge - hence the breakdown. That's what's actually utilized in thyratrons to switch high currents in tubes.
-
the Vce(sat) voltage was an interesting point that was brought up ...
How about using 3 IGBTs and one fet?
the fet can be used to short out the output as well as allowing for low voltage applications. It could also be connected to a 16 bit DAC combined with a matching ADC on a common shunt to get <1mA resolution going all the way to 20A.
IGBTs are then used for higher power applications to take the load off of the mosfet.
I've seen "Marco Reps" use a huge IGBT as a load and he didn't seem to have any problem with it either?
https://www.youtube.com/watch?v=acAuW0IVXKw (https://www.youtube.com/watch?v=acAuW0IVXKw)
-
My preferred method, switched resistors in parallel with a (much smaller) current sink. Cheap, needs much less heatsinking. Very robust.
Tim
-
You have mentioned that before, but it's not a dynamic load if it's used in that way! that's the main point you're missing tim. I want to build a dynamic dummy load with as much precision as I can.
-
How dynamic you want? I get >10kHz bandwidth with that method. :-//
Tim
-
You have mentioned that before, but it's not a dynamic load if it's used in that way! that's the main point you're missing tim. I want to build a dynamic dummy load with as much precision as I can.
In addition to poor SOA that make the IGBTs essentially useless to linear operation they are also slow - very slow compared to MOSFETs.
For the precision one might consider combining MOSFETs and BJTs in a kind a IGBT replacement circuit. This could reduce gate leakage, but a don'T think gate leakage would be a big deal anyway.
If high precision / good resolution is wanted it is more about the shunts / resistors. From an older electron microscope (needs quite a few constant current sources for the magnets, in this case some 6 channels of some 5 A at around 50 V) a remember a set of resistors of significant size: likely 50-200 W power rating wire would each. The resistors could be a more difficult (and expensive) part compared to the power semiconductors. It is not just the power rating needed, but also low TC if it needs to be accurate.
-
the absolute value of the shunt is not a big deal as I have access to high precision lab equipment in our uni so I can calibrate that out in software, I'd be more considered about the temperature stability of the shunts :D
And by dynamic load, I don't mean the frequency response, rather being able to have constant current,constant power. a simple resistor cannot do that.
In addition to poor SOA that make the IGBTs essentially useless to linear operation they are also slow - very slow compared to MOSFETs.
For the precision one might consider combining MOSFETs and BJTs in a kind a IGBT replacement circuit. This could reduce gate leakage, but a don'T think gate leakage would be a big deal anyway.
If high precision / good resolution is wanted it is more about the shunts / resistors. From an older electron microscope (needs quite a few constant current sources for the magnets, in this case some 6 channels of some 5 A at around 50 V) a remember a set of resistors of significant size: likely 50-200 W power rating wire would each. The resistors could be a more difficult (and expensive) part compared to the power semiconductors. It is not just the power rating needed, but also low TC if it needs to be accurate.
what do you think about my last idea? 3 IGBTs to take up most of the load, plus a linear fet for the added precision?
-
This IGBT is fairly cheap and seems fit for the job. It has a SOA chart unlike what previously has been said, and given the numbers I provided, it seems like a fairly decent choice.
https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?
(https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?qs=sGAEpiMZZMv4z0HnGdrLjmAnkg%2f6XuvzWCGzpW%2fDvWZ3y7weEXwbsA%3d%3d)
That's interesting, most IGBT's don't promise anything for DC SOA.
Fairchild's trench stop technology based IGBT's seem to have DC SOA specification, added benefit of trench stop or just hopefull specification?
I think the DC SOA shown is a lie. IR shows the same thing in their datasheets but in their application notes gives a warning that they are not characterized for linear operation.
Get some samples and test them to find out.
-
If you are building big 1000 volt electronic loads, then you do not need my help. You will require lots of devices in series and parallel to distribute the power and handle the worst case safe operating area no matter what power pass element you use. Those big linear MOSFETs from IXYS start to look like a deal if they cut the number of pass devices down.
At that high voltage, wouldn't tubes start making sense?
At higher voltages, I have seen EIMAC transmitter tubes used as power pass elements but their main advantage is ruggedness when overloaded and not cost. At those prices, linear rated power MOSFETs become competitive even when many devices have to be used in series and parallel.
-
I think the DC SOA shown is a lie. IR shows the same thing in their datasheets but in their application notes gives a warning that they are not characterized for linear operation.
Get some samples and test them to find out.
I highly doubt that the manufacturer is going to lie about their product details. I will get some for testing for sure, but I have high hopes for this device
-
some people mentioned BJTs as another way to dissipate the energy into heat. They are current controlled devices tho ... how could they be implemented in a similar fashion? can anyone give examples of a circuit that uses a BJT?
I would imagine they are probably cheaper and higher rated than the IGBTs ???
I might be able to use a FET for the precision and the BJTs for the bulk of the load
-
I think the DC SOA shown is a lie. IR shows the same thing in their datasheets but in their application notes gives a warning that they are not characterized for linear operation.
Get some samples and test them to find out.
I highly doubt that the manufacturer is going to lie about their product details. I will get some for testing for sure, but I have high hopes for this device
If they are not intended for linear applications, then why would they bother to destructively characterize them for it? And if they do not destructively characterize them for it, what are they going to show on the datasheet? I will not believe it until someone destructively tests a good sample size.
This application note (https://www.onsemi.com/pub/Collateral/AN1628-D.PDF) has some comments about SOA testing.
some people mentioned BJTs as another way to dissipate the energy into heat. They are current controlled devices tho ... how could they be implemented in a similar fashion? can anyone give examples of a circuit that uses a BJT?
I would imagine they are probably cheaper and higher rated than the IGBTs???
The big advantage of bipolar transistors is that excluding secondary breakdown, they will be the least expensive option for a given power level simply because they cost less per square millimeter of silicon area. Unfortunately secondary breakdown makes a big difference so linear rated power MOSFETs may be more economical especially now due to economy of scale.
-
I will not be pushing the devices as hard as I can. I will allow a lot of headroom, so the secondary break would not be a problem. Can you please provide schematics of how a BJT can be used instead of the mosfet? is it the same circuit with the output of the op amp connected to the base of the transistor?
-
some people mentioned BJTs as another way to dissipate the energy into heat. They are current controlled devices tho ... how could they be implemented in a similar fashion? can anyone give examples of a circuit that uses a BJT?
I would imagine they are probably cheaper and higher rated than the IGBTs ???
I might be able to use a FET for the precision and the BJTs for the bulk of the load
The company, TDI (Transistor Devices Inc, now astrodynetdi.com ?) has made dynamic loads (Dynaload) for many years and we have bought and used many of these from E-bay for the past 25+ years. They have different modes, constant voltage, constant current, pulsed, external control, etc....
They used TO-3 BJTs in parallel and emitter resistors, just like an audio power amplifier, for transistor sharing. Also good heat sink and fans for cooling. We have had to fix a couple of them and we would replace the BJTs with better parts like the On-Semiconductor MJ15024 parts or similar. The old ones at least used not very good BJTs like, 2N3055s as I remember. MJ15024 (and that general family) are 250V parts with a good SOA and we used these in audio power amps for many many years as well as other amplifier companies.
I didn't think you needed higher voltage ?
boB
-
I will not be pushing the devices as hard as I can. I will allow a lot of headroom, so the secondary break would not be a problem. Can you please provide schematics of how a BJT can be used instead of the mosfet? is it the same circuit with the output of the op amp connected to the base of the transistor?
It is the same circuit however because the base current of a bipolar transistor contributes to the emitter current, either a Darlington configuration or power MOSFET is used to drive the output transistor for better accuracy.
-
Some manufacturers are more honest with SOA curves than others. There a quite a few questionable to likely wrong FBSOA curves in MOSFET data-sheets.
A problem with the SOA curves in some data-sheets is that for switching applications it relates to the dynamic thermal resistance. So they measure / calculate the pulsed thermal response and than calculate the FBSOA from that. However this way is ignoring thermal instabilities and is thus valid only for short pulses. So the curves for 1 µs up to maybe 1 ms may be valid - but the longer pulse times and DC may not be real.
So if there is a pulsed thermal response curve (which is normal for more modern switching devices) one should be careful with the FBSOA, if it does not include the steeper roll off due to thermal instabilities.
Another problem is that not all samples may behave the same - so sample testing may not be good enough to get reliable devices. For this reason there are individually SOA tested BJTs for audio applications available and also linear rated MOSFETs may be SOA test. However this comes at a price.
-
How would you non-destructively test SOA? Motorola's old application notes on the subject only mention destructive SOA testing and specifically warn against trying to grade parts based on SOA for this reason.
-
Non destructive SOA testing can be tricky. My idea for at least the near DC case would be to run the DUT in forced constant current mode and carefully monitor the gate voltage. When instability starts I would expect to get a significant drop in gate voltage, faster than from the pure homogeneous warming. Still this is tricky and might cause some destructive events.
The other point is an SOA check, to see if the samples survive a give load pulse. This would be nondestructive for the good ones and destructive for the bad ones.
-
so if I intend to use a BJT, I should add a base resistor which is then driven by a darlington pair that is controlled by an OP amp? wouldn't that be more complicated than a straight up IGBT? I'm still not sure about the overal schematic ... like how much base current would I need? what value of base resistor should I use? where would the op amp feedback come from? It would be awesome if someone can post a picture of the schematic. Thanks.
-
I highly doubt that the manufacturer is going to lie about their product details.
Happens all the time, day in day out. Completely normal business practice, regardless of how highly you doubt it.
Yet, you have to trust something. Try it out.
-
I think the preferred way is to run increasingly longer pulses and stop when the pulse just starts to curl over. Then repeat that at different voltages and currents to sample points in the SOA plane.
Vgs drop is a tricky basis, because it's a microscopic part of the transistor that's hogging all the current and it doesn't take much change for it to burn through. That's probably the first thing you need to establish, how much droop or what rate of droop is acceptable. If neither varies with current, you may need a lot of transistors...
Tim
-
I think the preferred way is to run increasingly longer pulses and stop when the pulse just starts to curl over. Then repeat that at different voltages and currents to sample points in the SOA plane.
Vgs drop is a tricky basis, because it's a microscopic part of the transistor that's hogging all the current and it doesn't take much change for it to burn through. That's probably the first thing you need to establish, how much droop or what rate of droop is acceptable. If neither varies with current, you may need a lot of transistors...
Tim
This sounds like a better and practical method, though it might still be tricky to see the fine difference of just starting to become unstable before blowing up.
so if I intend to use a BJT, I should add a base resistor which is then driven by a darlington pair that is controlled by an OP amp? wouldn't that be more complicated than a straight up IGBT? I'm still not sure about the overal schematic ... like how much base current would I need? what value of base resistor should I use? where would the op amp feedback come from? It would be awesome if someone can post a picture of the schematic. Thanks.
The IGBT is normally just not a real option due to the very limited SOA - so a big ass IGBT might be no better than an TO220 MOSFET like IRF510.
The BJT solution could be the IGBT equivalent circuit, with a N-Channel MOSFET driving a BJT in a Darlington like circuit. A simple darlington would be another possible way, though slightly less accurate.
-
Can you please provide some sort of schematic?
I'm definitely going with the mixed method. A single mosfet for the high accuracy with 16 or 24 bit ADC and DAC (they're both fairly cheap) and some other linear device to take up the majority of the load. if a BJT or darlington could be used, I would appreciate if you could provide some sort of schematic as how to implement them.
-
I will not be pushing the devices as hard as I can. I will allow a lot of headroom, so the secondary break would not be a problem. Can you please provide schematics of how a BJT can be used instead of the mosfet? is it the same circuit with the output of the op amp connected to the base of the transistor?
Maybe Bob Pease could help you ;) (https://www.youtube.com/watch?v=2N6cjGS7lUE)
-
OM222O said "Can you please provide some sort of schematic?"
From an earlier EEvlog posting, https://www.eevblog.com/forum/testgear/dynaload-dlf-100-100-1500-(100v-100a-1500w-dc-load)-teardown-repair/ (https://www.eevblog.com/forum/testgear/dynaload-dlf-100-100-1500-(100v-100a-1500w-dc-load)-teardown-repair/)
there was one PDF with a schematic of a 100 watt TDI Dynaload... The large Dynaloads are very similar though from what I can tell.
boB
-
they're using mosfets tho ... I was asking about a BJT equivalent
-
they're using mosfets tho ... I was asking about a BJT equivalent
TDI uses (used) BJTs.
-
what topology do you use for extremely high frequency dynamic loads? whats the road map look like?
-
It's going to be used to test small DC power supplies and battery packs so it doesn't need crazy high frequency response ... even a few kilohertz would be sufficient. as I mentioned before my maximum ratings are 50 votls/20 amps/ 200 watts (maybe 250 to 300 watts if I can find a suitable cooling solution).
There isn't yet an exact "topology". I'll be using an ESP 32 so I can have wifi functionality and write testing software on my PC. there would be a mosfet with a 16 bit DAC connected to it's gate to get a really good resolution over the entire range (~1 mA) but the accuracy would be questionable so I have to test the circuit in action.
I'll also be using a "common shunt" with a 24 bit ADC(I don't know why ADCs are so much cheaper than DACs but I won't complain lol), but any other FET/IGBT/BJT would have it's own shunt connected to a much lower resolution (10 bit) ADC. This will allow for a course current adjustment on the parts that take up the majority of the load, but allows a fine adjust with the one FET. This entire project should cost less than 70 bucks and would be a pretty killer device for the average hobbyist.
so far I'm really liking the ADS1219 but haven't decided on the DAC. I also don't know which device is best for linear application ... BJTs should be the cheapest and most robust but I still haven't figured out a circuit diagram that works with them :D
There would be temperature monitoring for each individual part so thermal runaways are not a possibility.
-
I watched a few teardowns of the commercially available DC loads ... the BK Precision 8500 uses IRFP250N as the power mosfet. checking the datasheet, there's no SOA mentioned which makes everyone jump to the conclusion that they're not rated for linear applications ???
Obviously BK Precision doesn't just use random parts and have done their own testing to make sure that this mosfet meets their requirement. How can somebody know if the part is suitably rated without destructive testing?
The more I dig into this subject the more confused I am as to which part performs the best for a given price budget ...
-
what topology do you use for extremely high frequency dynamic loads? whats the road map look like?
What is extremely high frequency 1MHz, 100MHz, 10GHz?
If you go into the MHz your wirering begins to behave as transmission lines with impedances that are frequency dependant and a lot higher than dc resistance.
There it starts to become a whole different design.
I watched a few teardowns of the commercially available DC loads ... the BK Precision 8500 uses IRFP250N as the power mosfet. checking the datasheet, there's no SOA mentioned which makes everyone jump to the conclusion that they're not rated for linear applications ???
Obviously BK Precision doesn't just use random parts and have done their own testing to make sure that this mosfet meets their requirement. How can somebody know if the part is suitably rated without destructive testing?
The more I dig into this subject the more confused I am as to which part performs the best for a given price budget ...
I stumpled over this too.
You can bet a dollar they tested this and perhaps they have a test method to sort them.
If the part is not rated for linear mode this does not exclude it per se, the manufacturer did not test nor garanties a SOA at DC.
And as you can see, there are quite an amount of them in parallel that would suggest a higher power rating...
There are few modern mosfets with linear mode SOA in datasheet and they all suffer from not giving SOAs for other die temperatures than 25°C.
There is one exception I know: IXYS has a series of linear rated MOSFETS that give another SOA for more convenient 75°C (I hopefully remind correctly).
-
I watched a few teardowns of the commercially available DC loads ... the BK Precision 8500 uses IRFP250N as the power mosfet. checking the datasheet, there's no SOA mentioned which makes everyone jump to the conclusion that they're not rated for linear applications ???
It's going to be hard to beat in cost.
There's nothing else you need to shop for. Only power dissipation, with a full SOA, and cost.
FWIW, FQA9N90C is one of the best deals I've seen among high voltage FETs. Offhand, it's probably going to be hard finding anything with as big of a die (power rating correlates with die area) for as cheap.
Tim
-
I watched a few teardowns of the commercially available DC loads ... the BK Precision 8500 uses IRFP250N as the power mosfet. checking the datasheet, there's no SOA mentioned which makes everyone jump to the conclusion that they're not rated for linear applications ???
Obviously BK Precision doesn't just use random parts and have done their own testing to make sure that this mosfet meets their requirement. How can somebody know if the part is suitably rated without destructive testing?
The more I dig into this subject the more confused I am as to which part performs the best for a given price budget ...
Fairchild publishes DC SOA specs for their IRFP250 (http://pdf.datasheetcatalog.com/datasheet/fairchild/IRFP250.pdf).
-
Using a vacuum tube might still be attractive at high voltage, as they tend to fail open - while semiconductors tend to fail short. However the minimum working voltage is rather limited.
...
What about oldy BU208 and BU508 transistors? Linear 1500V TO3 package
The BU208 / BU508 are not that good SOA wise. The right MOSFET types (usually higher voltage types even for just 30 V use) are still the better choice at higher voltages. The BJTs (especially audio types) might be a good choice up to about 50 V, maybe 100 V)
never said that. please quote me correctly next time. ;D
-
Fairchild publishes DC SOA specs for their IRFP250 (http://pdf.datasheetcatalog.com/datasheet/fairchild/IRFP250.pdf).
I had a look ... they have a tiny SOA ... no wonder they used 10 of them for a 250 watt unit!
I know that the die area and power dissipation are correlated, which is exactly why I'll be using 2 devices under 1 heat sink. this will spread out the heat and allow them to run cooler for the same amount of load.
I looked at the IGBTs again and man ... they have ridiculously low thermal resistance!
FGH50T65SQD-F155 is rated at 0.56 degrees C/W ... that means a delta of only 28 degrees for my application under full load!(50 watts per device).
Any generic 95 watt CPU cooler should be able to handle them while keeping them under 70c! that's just crazy!
for some reason the mosfets that I'm finding are in the neighborhood of 1 to 1.5 degrees C/W which means 2 to 3 times higher delta values!
I'll order a few of these bad boys for testing. I'll publish the results later ;)
-
Fairchild publishes DC SOA specs for their IRFP250 (http://pdf.datasheetcatalog.com/datasheet/fairchild/IRFP250.pdf).
I had a look ... they have a tiny SOA ... no wonder they used 10 of them for a 250 watt unit!
I know that the die area and power dissipation are correlated, which is exactly why I'll be using 2 devices under 1 heat sink. this will spread out the heat and allow them to run cooler for the same amount of load.
I looked at the IGBTs again and man ... they have ridiculously low thermal resistance!
FGH50T65SQD-F155 is rated at 0.56 degrees C/W ... that means a delta of only 28 degrees for my application under full load!(50 watts per device).
Any generic 95 watt CPU cooler should be able to handle them while keeping them under 70c! that's just crazy!
for some reason the mosfets that I'm finding are in the neighborhood of 1 to 1.5 degrees C/W which means 2 to 3 times higher delta values!
I'll order a few of these bad boys for testing. I'll publish the results later ;)
FQA9N90C mentioned by Te51acOi1 has 0.45 C/W
Old generation high voltage parts with large dies are winner here.
-
you are correct, but look at the current rating! only 9 amps maximum!
If I want to make a large number of these products later (I intend to sell them if they meet my expectations), I can still have the same voltage and current rating, but just drop the power rating to match the cooling.In any case I'll buy a few of those as well. they seem like quite a nice option as well.
-
what topology do you use for extremely high frequency dynamic loads? whats the road map look like?
The next step up in dynamic performance uses a cascode output with a low voltage transistor driving the emitter/source of a high voltage transistor. This isolates the reverse transfer capacitance of the output transistor yielding better high frequency performance.
-
Fairchild publishes DC SOA specs for their IRFP250 (http://pdf.datasheetcatalog.com/datasheet/fairchild/IRFP250.pdf).
I had a look ... they have a tiny SOA ... no wonder they used 10 of them for a 250 watt unit!
I know that the die area and power dissipation are correlated, which is exactly why I'll be using 2 devices under 1 heat sink. this will spread out the heat and allow them to run cooler for the same amount of load.
I looked at the IGBTs again and man ... they have ridiculously low thermal resistance!
FGH50T65SQD-F155 is rated at 0.56 degrees C/W ... that means a delta of only 28 degrees for my application under full load!(50 watts per device).
Any generic 95 watt CPU cooler should be able to handle them while keeping them under 70c! that's just crazy!
for some reason the mosfets that I'm finding are in the neighborhood of 1 to 1.5 degrees C/W which means 2 to 3 times higher delta values!
I'll order a few of these bad boys for testing. I'll publish the results later ;)
If you're playing the low thermal-resistance game then check out some of the other Fairchild parts, e.g.:
http://www.onsemi.com/PowerSolutions/product.do?id=FGA60N65SMD (http://www.onsemi.com/PowerSolutions/product.do?id=FGA60N65SMD) (0.25 K/W J-C)
http://www.onsemi.com/PowerSolutions/product.do?id=FGY75N60SMD (http://www.onsemi.com/PowerSolutions/product.do?id=FGY75N60SMD) (0.2 K/W)
(also see FG60N60SMD for another case variant of the first one, and there's some others in the series as well, probably including cheaper diode-less versions)
These have thin (75 micron IIRC) dies with fairly large area to give those numbers. Note that while these _do_ give a DC SOA, they also specify that it's a "single pulse" (i.e. not really DC after all). Cheap though, so feel free to test one out to see if it lets out the magic DC linear region smoke.
I can confirm that they can take some abuse in switching though - these parts are well loved in the solid state tesla coiling scene (I've had the 75N60 hard switching >200A @ 350kHz, albeit at a low duty cycle)!
Edit:
Remember that a sil-pad or similar insulation could easily double the thermal resistance when dealing with stuff this low. In my use I've got isolated heatsinks to avoid having anything other than thermal grease in the way.
Also massive "brick" style IGBTs and power BJTs are available as surplus for reasonably prices. These will have rated dissipation in the kW region and are easy to mount with inbuilt isolation, but again, linear DC operation is _not_ what they are designed for, and the use of multiple dies internally will probably make any hotspot issues even worse.
-
do they ever use more advanced current mirrors and stuff like that? as inefficient was it would be
-
The only devices that are connected to each heat sink, would be the IBGT or MOSFETs which will have a common drain anyways. No need to use a pad :) just some high quality thermal paste. I have a lot leftover after I changed my laptop and xbox thermal paste. I think it's 4 watt per meter kelvin :P absolutely no issues there.
-
The only devices that are connected to each heat sink, would be the IBGT or MOSFETs which will have a common drain anyways. No need to use a pad :)
I thought we were talking high enough voltage that it needed devices in series, or is that two different conversations going on in here that I missed something? :)
-
The only devices that are connected to each heat sink, would be the IBGT or MOSFETs which will have a common drain anyways. No need to use a pad :)
I thought we were talking high enough voltage that it needed devices in series, or is that two different conversations going on in here that I missed something? :)
Yeah ;D there are two different conversations going on in here. I just mentioned that IGBTs usually have higher voltage ratings for similar price at the same current capability. however, I will be using multiple devices in parallel to do load balancing and increase the overall die area which helps with cooling.
-
Having the heat sink connected to the output terminal can be an issue, as it requires the heat sink to be well insulated from the case and thus well inside. It depends on the overall case if this is an option or not. The other point is the option of using individual fuses for the MOSFETs - these would normally be on the drain side.
Due to the SOA limitations one usually can only use a small fraction of the maximum P_tot. Thus the mounting on the heat sink is normally not that critical. I would consider something like 20 W for a TO220 case and 47 W for a TO247 case as a practical limit. In an electronic load there is no real practical protection to a higher than recommended voltage - thus for robustness its best to have some reserve, so that even a load intended for 50 V could withstand something like 100 V or 200 V at least for a short time until protection (turn off) can kick in. Another troublesome case is a large series inductor and thus possibly some energy to dump in avalanche mode as a result of an attempted fast turn off.
At least those old type 400-600 V MOSFETs are not that expensive. One might need a few more to get a low minimum drop and high current at low voltage anyway. The limited current rating is usually not an issue, unless one needs an electronic load for very low voltage use (e.g. single cell battery dicharge / testing or CPU simulator), it is more the SOA at higher voltage and R_on al low voltage that is limiting.
IGBTs usually make no sense for low voltage and also not for high power dissipation. So the IGBT is not a good idea :horse:. :horse:,....
-
If you read my earlier post, I intend to use high precision parts as well as a mosfet to gain that accuracy at lower voltages and a very precise current over the full 20A range as well as IGBTs to handle the majority of the load
-
If you read my earlier post, I intend to use high precision parts as well as a mosfet to gain that accuracy at lower voltages and a very precise current over the full 20A range as well as IGBTs to handle the majority of the load
Please read the other posts too, and forget about the IGBTs idea - essentially all of them are not suitable in the normal linear operating electronic load.
One can likely get away with just MOSFETs in the power stage. If needed it may help to have different size circuits for low and high currents, so one does not need to use the 20 A capably circuit for less than some 100 mA. A slight limitation of the normal MOSFET circuit would be there gate current, but this is usually a really tiny current.
For high precision the more difficult part would likely be the choice of the right shunt resistor - expect to spend more on the resistors than on semiconductors.
-
you are correct, but look at the current rating! only 9 amps maximum!
Well yeah, that was an example. Mine is higher voltage than yours --
(https://www.seventransistorlabs.com/Images/ActiveLoad1.jpg)
Read it as encouragement, to go shopping and find something effective and affordable, and at your power level I'd consider TO-220 as well as TO-3P and TO-247. :)
Tim
-
you should put 10 turn pots on that if your not selling it
also grounded chassis or plastic knobs
-
I'm mostly looking at TO 247 packages because they have more metallic surface area which has the least thermal resistance and they are physically larger too which means they're easier to cool. The power circuitry would be just a switch mode PSU (like the laptop chargers) to provide 12 volts input which is then split to other rails using LDOs. At these sort of voltages grounding the body doesn't really matter. even if the load line somehow shorts out to the body, it would be 50 volts max (voltage sense circuit will cut the current if over voltage is detected) so it's pretty safe. You can't even feel 50 volts DC for the most part. Honestly the exact value of current shunts doesn't matter at all. I have access to high precision tools and can measure each one and calibrate for them in the software. I just need high temperature stability and I'll make sure to put them in an area that they receive air from the cooler blowing down on them.
I ordered the IGBTs so it's too late to not try them out lol. I honestly don't see why people are so skeptical about using them, given that they have guaranteed DC SOA in the data sheet. The manufacturer doesn't warn that they're not intended for use in their linear region anywhere on the datasheet as far as I can tell ???
Anyways, I'll update you when I get the parts and do my testing on them. I ordered a few extras for the destructive testing purposes just to see how far I can push them with the cooling solution that I have and what external case temperatures might be dangerous as there is no way for me to directly read the junction temperature.
Just to make sure I also emailed ON semi technical support and described my applications and specifications and asked if there would be any problems running them in their linear region. If I get any answers, I will post them here.
-
well some IGBT require a negative voltage or at least recommend a negative voltage for switching off.
If it has a DC soa then it seems fine, unless there is something weird about IGBT sinking transient/AC while biased or bias change while under load.
-
you are correct, but look at the current rating! only 9 amps maximum!
Well yeah, that was an example. Mine is higher voltage than yours --
(https://www.seventransistorlabs.com/Images/ActiveLoad1.jpg)
Read it as encouragement, to go shopping and find something effective and affordable, and at your power level I'd consider TO-220 as well as TO-3P and TO-247. :)
Tim
Where's the heat sinks, Tim ? One per transistor ?
boB
-
I'm mostly looking at TO 247 packages because they have more metallic surface area which has the least thermal resistance and they are physically larger too which means they're easier to cool.
Ever consider SOT-227 body ?
My IXYS IXTN46N50L 500V VDSS 46A FET still sitting idle as I don't have time to build mine yet, planned to use a desktop oc-ing cpu heatsink.
(https://www.eevblog.com/forum/projects/(need-idea-ask)-securing-sot-227-mosfet-to-heatsink-base-mosfet_s-fbsoa/?action=dlattach;attach=225611;image)
(https://www.eevblog.com/forum/projects/(need-idea-ask)-securing-sot-227-mosfet-to-heatsink-base-mosfet_s-fbsoa/?action=dlattach;attach=225609;image)
-
Where's the heat sinks, Tim ? One per transistor ?
boB
Those are the switching channels (using same transistor for convenience, though a much smaller type would ideally be used), the linear ones are on the back row. I put together a bigass heatsink for them, two heatsinks strapped together, supported on fiberglass insulators and monitored with a thermistor. :)
Ed: this will make more sense: https://www.seventransistorlabs.com/Images/ActiveLoad2.jpg (https://www.seventransistorlabs.com/Images/ActiveLoad2.jpg)
Tim
-
is it actually worth using mesh around the fans? I thought that lowered their efficiency.
also if you made it longer and put the resistors lengthwise i think they would cool better
and there could be air flow through them. I think the change might be fairly drastic.
I always wondered about those resistors, if you can make fittings for them and connect them to a ducted fan.
-
I got an answer:
Thank you for contacting ON Semiconductor.
Unfortunately, it is not a good idea to use IGBTs in the linear region. Threshold voltage between individual pieces will vary considerably so if you set a working point for one IGBT and replace it with another, the performance will be different. For this reason it is difficult, if not impossible, to achieve consistency.
If you have any further questions, please do not hesitate to contact us back.
They're basically worried about the precision of the device if it's being ran from a fixed voltage value. In my case each IGBT is controlled from a separate op amp to get an accurate current through it ;D
They didn't mention anything about them not being able to handle DC current in their linear region and given the SOA provided in the datasheet, I'm farily confident that these are a better replacement than the mosfets
-
you should ask, maybe they thought to stop investigating or explaining the matter after the why not.
-
@OM222O --
you asked about the suitibility of IGBTs for operation in linear mode (as the pass element in an electronic load in particular). Many knowledgable members of the forum told you their thoughts and experiences. The answer was virtually unison that it's not a good idea to use an IGBT in linear mode, and this for a good reason. Of course, it's your own decision to try it and convince yourself. Maybe you find a solution that all the others didn't think of and succeed. But please don't try to argue against the leading opinion without solid evidence. A good point to start at is reading about failure modes of MOSFETs since they are very closely related to IGBTs (the latter have one doping layer more). Here's a book by NXP (https://www.nxp.com/docs/en/user-guide/MOSFET-Application-Handbook.pdf) that you may want to have a look at and that covers the subject quite well and provides informaton to other literature. Some interesting information can be found on page 43ff.
It's also for a good reason that most of the entry level electronic loads utilise the ancient IRFP250 (non-N version if possible) as their pass elements, and usually many of them, each with its individual driving circuitry. There have been many clever people scratching their heads how to solve this problem as elegant as possible. And it's rather smart to learn from them as much as possible and not to release too much magic smoke that you've got to pay for yourself...
Just my two cents.
Good luck and all the best,
Thomas
-
Mosfets will also have a difference of Vgs threshold and it changes with temperature. Even if you match your FETs or IGBTs there is no guaranty that they will track precisely over temperature range.
I'd use BJTs
You might get away with using insulated gate parts if it is only one and not more than one in parallel.
This is where switching and PWM with a load resistor can come in handy. Then the Vgs doesn't matter so much.
-
load matching transistors in a dynamic circuit by active means is just fugly.
-
Taking Vgs(th) out of the equation by attaching a large source resistor, or single op-amp, to each transistor, is not fugly, it's traditional.
They didn't do it back in the BJT days because, well, actually, they did, with emitter resistors, but op-amps no, because ICs were still expensive back then. Since the 70s passed, no one cares. ;)
The only non-integrated device in existence, that is well enough matched to not need a common-terminal resistor, is the vacuum tube. And even back then, they still recommended it! (Example: 6AS7/6080 regulated supplies, designs abound. The cathode resistor was more to limit short circuit current though, which on beefy tubes like these, could be several amperes!)
Tim
-
I meant digital adjustment.
-
This IGBT is fairly cheap and seems fit for the job. It has a SOA chart unlike what previously has been said, and given the numbers I provided, it seems like a fairly decent choice.
https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?qs=sGAEpiMZZMv4z0HnGdrLjmAnkg%2f6XuvzWCGzpW%2fDvWZ3y7weEXwbsA%3d%3d (https://www.mouser.co.uk/ProductDetail/ON-Semiconductor-Fairchild/FGH50T65SQD-F155?qs=sGAEpiMZZMv4z0HnGdrLjmAnkg%2f6XuvzWCGzpW%2fDvWZ3y7weEXwbsA%3d%3d)
(https://i.imgur.com/iunNkOC.jpg)
Can you please confirm I'm not making a mistake in reading the datasheet? It has way better SOA for similarly priced mosfets.
(Please keep in mind I'm limiting the power of each part to 50 watts, so @ 50 volts 1A would be the current limit and as you decrease the voltage, the current can go up accordingly, so my intended area of operation falls way below the provided SOA)
Well, there it is, plain as day! It can handle 100 A with just 2.5 V across it, but if it has 100 V across it, the SOA is only good to about 2 A. And, presumably that is at 25 C, if you let it get hot, the SOA probably shrinks.
Jon
-
Well, yeah.
Well, yeah.
Well, yeah.
That's precisely how SOA and thermal resistance works! Gold star! :P (Not to be condescending. Just that, alas, if only so many commercial designers understood it so simply...)
250-270W (by my eye, but anyway the RthJC figure is what's at work here) is no slouch of a transistor, if it's affordable at least (and the DC curve is actually real). 500W+ transistors are easy to find, but it's harder to find the cheapest watts per dollar (at a given rating), which is really all that matters here. :)
Tim
-
@OM222O --
you asked about the suitibility of IGBTs for operation in linear mode (as the pass element in an electronic load in particular). Many knowledgable members of the forum told you their thoughts and experiences. The answer was virtually unison that it's not a good idea to use an IGBT in linear mode, and this for a good reason. Of course, it's your own decision to try it and convince yourself. Maybe you find a solution that all the others didn't think of and succeed. But please don't try to argue against the leading opinion without solid evidence. A good point to start at is reading about failure modes of MOSFETs since they are very closely related to IGBTs (the latter have one doping layer more). Here's a book by NXP (https://www.nxp.com/docs/en/user-guide/MOSFET-Application-Handbook.pdf) that you may want to have a look at and that covers the subject quite well and provides informaton to other literature. Some interesting information can be found on page 43ff.
It's also for a good reason that most of the entry level electronic loads utilise the ancient IRFP250 (non-N version if possible) as their pass elements, and usually many of them, each with its individual driving circuitry. There have been many clever people scratching their heads how to solve this problem as elegant as possible. And it's rather smart to learn from them as much as possible and not to release too much magic smoke that you've got to pay for yourself...
Just my two cents.
Good luck and all the best,
Thomas
Well I haven't heard any "good reasons" as why not to run an IGBT in it's linear region. People have mentioned linear SOA which as I showed in the picture of the data sheet matches my requirements and even after asking the manufacturer, they are more worried about matching Vgs more than anything else. Please let me know if there's any other "good reasons" not to do so. I'm just trying different things to maybe make something better than what "everybody else does". If I can make this work, it's gonna be a killer deal in terms of electronic loads.
I still haven't managed to find a decent circuit that uses BJTs :-DD :-DD
maybe the same arrangement as the mosfet works fine with an addition of base resistor , but most schematics say beta matters a lot so they use darlington or even 3 BJTs connected to each other. They don't provide any reason as to why tho ???
any comments on that?
I have also considered PWMing a few load resistors as Tim mentioned but it won't be true constant current and I don't think iy's gonna be as accurately adjustable as I want it to be (24 bit PWM is not available anywhere that I could find and it's probably gonna cost an arm and a leg if it was), so I'm brushing it to the side for the time being.
-
The problem with DC SOA curves for parts that are clearly intended for switching applications is that it may be plain wrong. In this case chances are very high the SOA curve is only calculated from the transient thermal resistance using a formula that is no longer valid beyond some 10-100 µs as it ignores thermal instability - exactly the effect that limits the useful SOA in linear operation.
It is quite common to find this type of mistake for modern MOSFET data-sheets from some manufacturers - chances are hight they use the same faulty software to calculate the SOA curve for IGBTs too.
It takes very special IGBTs to have a useful FBSOA - so it is much more likely the SOA curve is a mistake than the IGBT actually useful for linear operation. Unless they explicitly mention linear operation one should not trust an DC SOA that just reflects P_tot: not for a modern low voltage MOSFET (< 400 V), not for an BJT and the least for an IGBT.
Internally IGBTs are build a little like a MOSFET providing the base current to an BJT. BJTs are known to have limitations due to 2 nd breakdown, from a thermal instability. It takes rather special BJTs to get a good SOA beyond 100 V. At the same power level IGBTs are using even smaller dies than BJTs and thus tend to be even more susceptible to thermal instability. So it is obvious that the shown SOA can not be true, at least for the higher voltages like >100 V. It up to you to decide if you trust is up to 5 V or 10 V or maybe 15 V.
If something sound to good to be true, consider the possibility that it is not true.
-
is it possible to drill out the package and add a little germanium window or something so that you can measure temperature and make your own SOA curve?
Or measure the die even faster for feedback. Is it on the surface or something burried where the overheating happens? Can you shorten response time?
-
I assume thermal instability comes from inadeqote cooling. like how you get more performance and less power leakage on CPUs and GPUs if you cool them properly. I think it was somewhere in the ball park of 4% power decrease for every 10c drop in temps. I assume other semiconductors work similar to that ???
and as I mentioned the calculated temperature in my application would be at high 70s or mid 80s in worst case scenario which is far below the provided SOA, even considering the de-rating factor :-//
please correct me if I'm wrong
-
I assume thermal instability comes from inadeqote cooling. like how you get more performance and less power leakage on CPUs and GPUs if you cool them properly.
...
please correct me if I'm wrong
If you don't mind, please have a look at this publication (https://www.google.de/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjVhp7zxqjyAhV9hP0HHYTcAzUQFnoECAMQAQ&url=https%3A%2F%2Fwww.richardsonrfpd.com%2Fdocs%2Frfpd%2FMicrosemi%2520How%2520to%2520Make%2520Linear%2520Mode%2520Work.pdf&usg=AOvVaw1h2YiBkiOeD7eYBAGT5suj) - it explains a lot and also points out possible solutions.
Edit: Broken Link fixed
-
Well, there it is, plain as day! It can handle 100 A with just 2.5 V across it, but if it has 100 V across it, the SOA is only good to about 2 A. And, presumably that is at 25 C, if you let it get hot, the SOA probably shrinks.
Not sure if serious ...
For the reasons I gave earlier, I do not believe the DC curve on that SOA graph. At a case temperature of 175C, the SOA graph is a point at zero volts and zero amps. I am kind of suspicious that they do not show a graph of this but I suspect it is either because they never intend this device to be used in linear operation where it would matter or because it actually performs worse than the specifications indicate.
I assume thermal instability comes from inadeqote cooling. like how you get more performance and less power leakage on CPUs and GPUs if you cool them properly. I think it was somewhere in the ball park of 4% power decrease for every 10c drop in temps. I assume other semiconductors work similar to that ???
Thermal instability comes from operating at high Vds where the temperature coefficient of the Vgs reverses. Secondary breakdown comes from operating at high Vce where current crowding causes localized heating raising hfe. So both effects are temperature related in that they cause thermal runaway through positive feedback.
and as I mentioned the calculated temperature in my application would be at high 70s or mid 80s in worst case scenario which is far below the provided SOA, even considering the de-rating factor :-//
The junction temperature directly affects the SOA and it affects the secondary breakdown and thermal instability regions of the SOA just as much if not more.
I'd use BJTs
His requirements are not particularly demanding:
I'll most likely aim for 50 volts, 20 amps, 200 watts and spread the 4 into 2 groups of 2 (again for increased surface area and better cooling), each cooled by a CPU cooler, so each device needs to handle roughly 50 watts which is not that large.
I suspect a single power MOSFET driving 4 bipolar transistors in parallel (or 8, see below) with 0.1 to 0.22 ohm emitter ballast resistors would be the least expensive because no extra effort would be needed for current sharing due to consistent Vbe and low emitter ballast resistance. The bipolar transistors could be purchased in a lot of 10 and matched for Vbe at 5 amps for additional reliability.
MOSFETs or IGBTs may be only slightly more expensive but may need to be driven individually to enforce current sharing. There are more choices now for big MOSFETs and IGBTs versus bipolar transistors contributing toward closer prices. It seems the only big non TO-3 bipolar transistors you can buy now are intended for audio. Conceivably an even less expensive design using the largest TO-220 bipolar transistors but more of them (2 per heat sink?) is possible.
I checked a couple of big power MOSFETs and except for the linear rated IXYS IXTH80N075L2 and IXTH64N10L2, the ones I checked would all have problems with current sharing. You can see it in the graphs where the ones intended for switching operation have transconductance increase with temperature while with the IXYS parts, it decreases. Figures 1 and 2 for that IGBT show the same thing but even worse. The switching MOSFETs also showed straight DC SOA which I do not believe but I do believe it on the IXYS parts. Who else makes linear MOSFETs?
-
You can argue all you want, but it only ever comes down to this:
What is the component capable of?
It doesn't matter if MOST IGBTs have a shity SOA. I don't trust that this one has a full SOA, so I would test it.
How do you test it? Build your circuit and see if it works! Duh! Simple as that! :D
There are many MOSFET types out there. A lot have 2nd breakdown within the operating area. You can't generalize "MOSFETs are free from 2nd breakdown", that used to be true back in the bad old days but it isn't now.
You can't generalize "IGBTs have shit SOA", though until this thread I hadn't seen one with a DC SOA, and so it would seem this generalization is more reasonable, but again not an actual rule.
You can't generalize "BJTs exhibit 2nd breakdown [in their operating area]" (note the often omitted but necessary qualification!), because while a lot of them do, amplifier transistors are made to handle it. Just as modern (high current density) MOSFETs are sometimes made in linear flavors, or coincidentally happen to have a full SOA.
Tim
-
Who else makes linear MOSFETs?
Microsemi does it, although not as many as IXYS/Littelfuse has.
-
I assume thermal instability comes from inadeqote cooling.
No. The thermal instability discussed here has little to do with external cooling - it's about local hotspotting, which happens so quickly and at microscopic level that it's all about the internal heat spreading of the die, through the substrate, to the device case. You can't do much externally.
This is, as has been explained numerous times, due to the different threshold voltage of the different parts of the transistor, and their positive tempco, meaning that a local part that takes more current through it, starts getting even more current, and runs away thermally.
This is actually the same discussed problem of Vgsth matching, but you can't do anything since it's happening inside the transistor - and they have done their best to minimize the source (emitter) resistance. Can't add separate amplifiers or emitter resistors there!
This effect is rolled into the SOA, and if your part has a DC SOA graph like this one, and it's OK for you, then go ahead. (Although it would be still quite difficult to be in-spec if they only give you a DC SOA for Tc=25 degC. You need a refrigerator cooling, basically. Or you are going outside the specs and are on your own anyway.)
Otherwise, the manufacturer is outright lying, or have no idea about the basics of their field of market.
There is only one problem, and it's the fact you can't believe a manufacturer could actually lie about their parts - or if not outright lie, at least not understand their own parts and what the basic specifications means. When you gain more experience, you'll find out the harsh truth. They lie all the time. And the datasheet writers are not always the brightest geniuses who actually make the transistors. Wrong info on the datasheets is so common that it isn't even a pet peeve of mine anymore - we just need to live with it. It's like a rainy day, we don't even feel like complaining about this would help.
This being said, the chances are the DC SOA is perfectly valid. Go for it.
Vgs (Vge) threshold matching is completely a red herring. There is zero need to match this parameter. Any sane design today uses separate drive per transistor, completely eliminating this parameter. Not using a separate driver is an ugly hack, only there to try to "save cost", which made sense when opamps did cost something. Now they come in miniscule packages and cost a few cents per amplifier. Now, not using dedicated drives only increases cost as you probably need to derate the transistors more, add more source(emitter) resistance, handpick parts or do some other hacks which have their own hidden costs (most likely much more than an opamp).
-
i found this quite helpful in understanding the derating of continuous DC SOA at elevated temperatures (PDF download):
https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200 (https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200)
-
To satisfy my curiosity I played with my lab power supply and 2pcs IRG4PC50W IGBT for SOA:
1. exemplar: 100V and 1 amp =immediate destruction
2. exemplar 100V and 1 amp = survived for test duration (1 seconds :-DD )
2. exemplar 200V and 0.5 amp = instant death
All of these were performed at about Tcase ~25c (200W "rated" power for these)
Pretty much as expected, unfortunately didn't have any DC SOA rated IGBT's around in my parts bins.
-
To satisfy my curiosity I played with my lab power supply and 2pcs IRG4PC50W IGBT for SOA:
1. exemplar: 100V and 1 amp =immediate destruction
2. exemplar 100V and 1 amp = survived for test duration (1 seconds :-DD )
2. exemplar 200V and 0.5 amp = instant death
All of these were performed at about Tcase ~25c (200W "rated" power for these)
Pretty much as expected, unfortunately didn't have any DC SOA rated IGBT's around in my parts bins.
You mean the case was 25 C before you applied power to it ?
No heat sink ?
-
Well I haven't heard any "good reasons" as why not to run an IGBT in it's linear region. People have mentioned linear SOA which as I showed in the picture of the data sheet matches my requirements
OK, the reason for all this is that the output stage of the IGBT is a bipolar transistor. These have a negative thermal coefficient of voltage drop, so the HOTTEST part of the die carries the most current. When paralleling multiple bipolar or IGBT transistors, you use ballast resistors to balance current. Within a single die, you can't do that. The IGBTs are designed to minimize this effect in full saturation, so the SOA gets much bigger when saturated. The ability to keep all parts of the die carrying the same current fails in the linear region, so one area of the die gets hotter and takes most of the current. That's why your 100 A transistor can only handle 2 A with higher C-E voltage across is.
And, that graph is for a case temp of 25 C! That is important. At higher temperatures, this effect can get a lot worse. At 100 C case temp, this transistor may only be safe at 100 mA.
Jon
-
You mean the case was 25 C before you applied power to it ?
No heat sink ?
Large 8mm thick copper plate as heat sink.
-
can you get lucky with luck of the draw for high IGBT linear current or just by their nature their gonna give you <3% of their rating in linear region performance?
does anyone have diagrams of these hot spot formations or thermographs?
what can I study to understand the charge densities and everything to understand the structure of the hot spot formation and why it decays with saturation.
-
There can be some scattering in when the parts will actually fail. However there is little chance to find a part that is much better - more like a small chance to find some early failing parts from even for types that normally have a good SOA.
The process of thermal instability formation can be described the following way:
We assume a given local heating to start the process, that will cause a temperature peak. The higher temperature will concentrate (in a device prone to instability) the current to that hot spot which than causes additional local heating. If this extra local heating is stronger than the initial heating assumes in the process, the system will likely run away to destruction unless possibly other nonlinear effects to stabilize it at a higher temperature state. So the thermal instability will not need some initial faults or inhomogeneity to start - it is more like feedback in an oscillator, one high enough it will start to oscillate (and is usually limited in amplitude by nonlinear effects).
So no chance find some magical much better samples, more like a chance to have some faulty ones that at some point's are even worse.
There is not magical limit to always allow some < 3% or similar in the linear region. It depends on voltage (higher voltage is usually worse) and the device type.
When going towards saturation there are 2 effects that help. One is less heat from a given increase in current and the second is that in many devices there would be no more current localization as the TC changes sign.
-
I ran across this paper when searching for "electro-thermal instability IGBT" https://www.nrel.gov/pv/assets/pdfs/2015_pvmrw_131_das.pdf (https://www.nrel.gov/pv/assets/pdfs/2015_pvmrw_131_das.pdf) Pictures on p.29 shows localized heating.
There are quite a few papers on the subject although not too many address linear operation specifically.
Cheers,