-
Dirt cheap and simple scope-based component tester - curve tracer
Posted by w2aew on 09 Jun, 2012 19:55 -
<shameless self promotion
>
I posted a new video that shows an old idea for simple component testing / curve tracing using an oscilloscope and just a few components. Circuits of this type have literally been around for decades, and are often referred to as an Octopus component tester. I really don't know the origin of the name. You can Google various search terms like octopus component tester curve tracer, etc. and find dozens of variants.
The video basically shows how this simple technique works, and demonstrates with some examples. Enjoy:
Alan - W2AEW -
I love your videos! I had never heard of this circuit before.
-
Nice video, thanks.
I first built an 'Octopus' whilst at Maritime College in the late 1980's. As you say, they have been around a long time and are often home built as so few parts are needed and they work so well on semiconductors.
Whilst working for my employer I was trained to use the Huntron Tracker instruments. For those who don't know, these are an 'Octopus' combined with a scope display to produce an all-in-one instrument for testing components. The Huntron units add a little to the design by having several voltage/current ranges to suit differing scenarios. They are very useful bits of kit, especially when hunting a fault on a PCB and you have a known good PCB as a reference to compare to. I bought a couple of used Huntron HTR1005B-1S a few years ago at a very reasonable price. Prices have sky rocketed since then so the word must be out that these are good units.
The Hameg oscilloscopes used to include an 'Octopus' mode in their design, named a Component Tester mode (CT for short). I have always thought them forward thinking to have done so. You can often pick up an HM203 series 20MHz CRO for a very low price or even free if you are lucky... if you do, you get a simplified version of the Huntron Tracker included for free !
Tektronix used to sell a advanced 'Octopus' I/V component tester under the model no. TR210....it was in fact made by Huntron and also sold under their model no. HTR-200. The unit was designed to connect to the TDS200 series DSO's in order to provide a component testing capability. It was far more advanced than the simple Octopus, HTR-1005B-1S or Hameg CT. I managed to pick one up a couple of years ago quite cheaply. If you see one at a decent price, you should consider buying it as it is a very refined piece of test kit with special low voltage/low current modes to suit modern low voltage MOS technology.
A word of warning..... Huntron carried out tests on TTL and CMOS technology to prove that the 50V test pulse that the Tracker could generate would not destroy the component under test. Huntron advise the use of the lower voltage ranges, but succeeded in proving that the chips tested still worked. It is interesting to note that Huntron later released the 2000 series that had more ranges and lower test voltages but these are still considered risky if used on very low voltage MOS components. The TR210/HTR-200 and later models all provide very low test voltage modes for VLV MOS and even have a special range lockout function to prevent accidental activation of the higher voltage ranges. Me thinks Huntron are not confident that the latest low voltage technology will take kindly to the older Huntron/Octopus excitation voltages. Be careful regarding this point if you DIY an Octopus for use on sensitive low voltage MOS technology.
The topic of this thread is 'Dirt Cheap' and none of the units I have mentioned, except the Hameg scope, could be considered cheap. A chap in the USA decided to build and market a cheaper solution. Take a look here:
http://www.actracer.com/
Whilst the unit is cheaper, I do not consider it cheap per se considering the component count but then this isn't a China made unit The ACtracer has a little bit more versatility than a basic Octopus and may be worth considering if the budget will stretch to it. Its certainly a lot cheaper than a new or used Huntron 2000 which it is trying to emulate.
Aurora
-
My Elektor (Heathkit) Transistor curve tracer, from 1980 I think, was my favourite test tool by far !!
I had the Hamegs, but they were no where near as nice. Being able to drive a transistor to destruction,
observing failure mode, then just backing off to recover ... multiple traces, sure miss it. -
Nice video, thanks.
<<<snipped>>>
Aurora
Aurora - awesome post, great information! I was aware of the Huntron units, and the Hameg scope with the CT option. That is similar to the Heath scope that I showed in my video, and I've also seen the same scope under other names. However, I did not know about the Tektronix/Huntron unit - and I work for Tektronix! Of course, I didn't for for them 14 years ago when this was a product. Very interesting!
Thanks for watching and posting. -
I am glad my posting was of interest.
If anyone wants to see the user manual, service manual or schematics for the TR210 I will be happy to provide them as I have the full set. Huntron provided the schematic and offer excellent support for their older products.
As a sort of 'spin-off' from the Huntron Tracker, Polar Instruments produced a component tester that offered very similar capabilities PLUS a curve tracer function for transistors. I bought it out of curiosity as I use Polar Tone Ohm's and had not seen the component tester before.
-
Interested in that, I can build it into a old Tek frame I have. The scope side works, but the medical recorder it had died a long time ago. Was looking to do this for a while.
-
Digsys,
Not quite the Elektor unit that you may have had, but take a look here:
http://www.ebay.co.uk/itm/Power-Transistor-Curve-Tracer-adapter-XY-Oscilloscopes-/140547987228?pt=BI_Oscilloscopes&hash=item20b950171c
The curve tracer is in fact a copy of an Elektor design with minor mods and is so cheap that I bought two to experiment with. I have not had time yet but you may want to buy one for experimentation as well, its certainly cheap enough
The schematic is provided and I have attached it to this message. The original 1989 Elektor design from which it was taken is available on the Internet.
I attach it as an advertisement for Elektor but will remove it if asked. Its is VERY old though
Elektor has a new Curve Tracer design, as can be seen here:
http://www.elektor.com/magazines/2009/february/transistor-curve-tracer.810360.lynkx
Aurora
-
You can also do this with a PC and a simple hardware interface that buffers the sound card.
Here are 2 articles on the subject published by Dr George R. Steber, WB9LVI, they cover the same ground but one goes more into the theory the other is about the hardware.
The 3rd file is from a user of the earlier design, I include it as he provides a PCB layout for those who wish to take it beyond the breadboard.
Cheers
Chris
Edit:
The software is here: http://www.arrl.org/files/file/QEX%20Binaries/06_July/7x06Steber.zip
Works on WIN 7 64 Bit and WIN XP, I have not tested on WIN 8 but it should work.
-
Thanks.
-
Here is some more information on using I/V curve analysis units on components.
I also attach the Huntron US Patent for the original 'Tracker' design.
The schematics and service manuals for many of the Huntron Trackers are available FOC from several sites in the internet.
Aurora
-
Quote
Not quite the Elektor unit that you may have had ...
Thanks. I have often searched for the original kit from ~1980. It came with nice sloping front metalwork and nice decals.
The PCB also had a lot more parts, took me ages to assemble, if my memory is any good. Elektor were so ahead of their time.QuoteElektor has a new Curve Tracer design, as can be seen here
Saw that, not as impressed. Maybe I'm too cynical these days :-) -
This is a great video for beginners. And I just watched another four or five. Your videos go right to the heart of the questions that have been bugging me and strike just the right mix of theory and demonstration. Please keep them coming. I just watched "Use a scope to measure the length and impedance of coax" and with it I feel I much better understand characteristic impedance, termination, etc. Thanks!!
-
This is a great video for beginners. And I just watched another four or five. Your videos go right to the heart of the questions that have been bugging me and strike just the right mix of theory and demonstration. Please keep them coming. I just watched "Use a scope to measure the length and impedance of coax" and with it I feel I much better understand characteristic impedance, termination, etc. Thanks!!
Ken - thank you! I'm glad I'm hitting the "sweet spot" with you. Please feel free to shoot me a message with ideas for video topics that you'd like to see.
Alan -
IET also made a curve tracer. It was the STS-1600.
The display was LCD dot matrix, and it is rather slow to update, but it does work.
-
Love the video - it was so easy to follow along. Really great stuff.
Is there any reason why I can't use a function generator to generate the test signal? -
George,
If you take a look at the "ACTracer FG" webpage link that I provided earlier in the thread, you will see that the designer of that unit has added the capability of an external function generator as the excitation oscillator. The ability to change frequency is useful when testing capacitors of varying values.
Aurora -
The Hameg oscilloscopes used to include an 'Octopus' mode in their design, named a Component Tester mode (CT for short). I have always thought them forward thinking to have done so. You can often pick up an HM203 series 20MHz CRO for a very low price or even free if you are lucky... if you do, you get a simplified version of the Huntron Tracker included for free !
I'm happy to say that the current series Hameg Scopes still include the component tester as standard as can be seen on my HMO724.
Cheers,
Wayne.
-
Love the video - it was so easy to follow along. Really great stuff.
Is there any reason why I can't use a function generator to generate the test signal?
George - in this very simple Octopus design, the only reason that you can't directly use a function generator is because the scope's Ground reference is not connected to either side of the signal generator (the transformer in this case). On a function generator, the output is generally earth referenced. Of course, you could use a transformer at the output of the function generator.
To tell the truth, I didn't have a 6.3V transformer as shown most online Octopus schematics, and I was playing with different test frequencies, so I was really using a my function generator to drive the secondary of the 12V transformer that is shown in the video, and the primary was driving the simplified circuit. Thus, I was able to use very little drive from the function generator, since the transform stepped the voltage up. I simply adjusted the amplitude on my generator as needed to get the output voltage I wanted. The transformer provided the earth isolation so that I could use the junction of the DUT and current limit/sense resistor as the ground reference. -
Hmm, while I'm considering buying a filament power supply I'm trying to think of a way to use something I either already have or buy something more financially appropriate. I'd love to build this. Although I would rather not have an 8lb iron core step down transformer to live with it. I mean in reality, if I had a filament transformer I'd rather build a tube amp to put it in. So my considerations would be something like this perhaps... This is a 220V input 12V AC output transformer, but wouldn't the output simply be half of whats rated? 110V ac input you'd get 6V AC out right? I Googled octopus component tester and what I found asks for 3 volts off the center tap of a 6V ac transformer. I'm just not sure what values or component I'd need to tweak in order to conform the circuit to suit the power supply. I'm not dead set on this power supply. I just thought there may be something out there over the shelf that might be a better choice. Johnny W
http://www.ebay.com/itm/20W-Max-220V-to-12V-AC-0-086A-Home-LED-Light-Power-Supply-Electronic-Transformer-/400303095053?pt=LH_DefaultDomain_0&hash=item5d33ec7d0d -
Hmm, while I'm considering buying a filament power supply I'm trying to think of a way to use something I either already have or buy something more financially appropriate. I'd love to build this. Although I would rather not have an 8lb iron core step down transformer to live with it. I mean in reality, if I had a filament transformer I'd rather build a tube amp to put it in. So my considerations would be something like this perhaps... This is a 220V input 12V AC output transformer, but wouldn't the output simply be half of whats rated? 110V ac input you'd get 6V AC out right? I Googled octopus component tester and what I found asks for 3 volts off the center tap of a 6V ac transformer. I'm just not sure what values or component I'd need to tweak in order to conform the circuit to suit the power supply. I'm not dead set on this power supply. I just thought there may be something out there over the shelf that might be a better choice. Johnny W
http://www.ebay.com/itm/20W-Max-220V-to-12V-AC-0-086A-Home-LED-Light-Power-Supply-Electronic-Transformer-/400303095053?pt=LH_DefaultDomain_0&hash=item5d33ec7d0d
There's really nothing set in stone with this. Using a 220VAC - 12VAC transformer on 110VAC will indeed give you 6VAC out. It's really up to you to decide how much AC voltage you want to use as your excitation voltage. You can always use a voltage divider at the output to cut it down further, as most of the Octopus schematics show. You can even use an audio isolation transformer on the output of a function generator if you want. You are also free to choose what current limit/sense resistor you want to use. My simplified example was used to simply show the concept. -
Alan,
Thanks for the very informative / educational videos - You've got a new Youtube subscriber.
I've tried your curve tracer with my Iwatsu ss-5705 today, it didn't work as expected. Sorry for my n00b question - is it because CH1 and CH2 are completely out of phase? I try to locate an 'invert' switch on the scope, but couldn't see anything. Does it mean that I have to build an inversion circuit to use my scope as a curve tracer?
Thank you in advance.
I've attached some pics:
Wave forms of CH1 and CH2 when connected to the 'Octupus'
X-Y Mode is on:
Measuring a capacitor
Measuring a diode
-
Really no need to build an inverting circuit. Just recognize the fact that +current through your DUT is deflecting in the downward direction instead of up. You're waveforms look OK, there's some phase shift beyond the 180 degrees due to the non-inverted ch2, but that could be inherent to the scope, and I wouldn't sweat it. Looks like your scope doesn't have an Invert function on CH2 because it includes a Ch1-Ch2 feature. Many other analog scopes have only a Ch1+Ch2 (ADD) feature, so the Ch2-INV was added so that you could also do the subtraction.
-
I've tried your curve tracer with my Iwatsu ss-5705 today, it didn't work as expected. Sorry for my n00b question - is it because CH1 and CH2 are completely out of phase? I try to locate an 'invert' switch on the scope, but couldn't see anything. Does it mean that I have to build an inversion circuit to use my scope as a curve tracer?
Make sure that you have both inputs DC coupled. That should take care of the problem. You can invert the Y axis by pulling the channel B position control out (it's labeled PULL INV).
-
I've tried your curve tracer with my Iwatsu ss-5705 today, it didn't work as expected. Sorry for my n00b question - is it because CH1 and CH2 are completely out of phase? I try to locate an 'invert' switch on the scope, but couldn't see anything. Does it mean that I have to build an inversion circuit to use my scope as a curve tracer?
Make sure that you have both inputs DC coupled. That should take care of the problem. You can invert the Y axis by pulling the channel B position control out (it's labeled PULL INV).
Excellent point about DC coupling - and nice catch on the pull-inv function - I missed that in the photos. Sounds like you've got him all sorted now.