Hi Clem,
You only really need to test the beta at one point. If it's out of range for what is normal, test another. If it's out - run.
KISS. Keep the tests simple and fast. After testing some "real" samples, you will get an idea where the range is. If it's simple to test, other DIY'ers can make the same jig and share the results. In this way, we all get a good basis of information.
Arif likes destructive testing.
-Chris
You only really need to test the beta at one point. If it's out of range for what is normal, test another. If it's out - run.
KISS. Keep the tests simple and fast. After testing some "real" samples, you will get an idea where the range is. If it's simple to test, other DIY'ers can make the same jig and share the results. In this way, we all get a good basis of information.
Arif likes destructive testing.
-Chris
Curmudgeon said:
Large dies are not cheap at all. I have never seen or had news concerning any fake device featuring a die larger than genuine one. That's why capacitance measurement is so powerful.
All the fakes that I've stumbled into show half the capacitances of the original devices or less. That's particularly true for Motorola/On-Semi devices since the genuine ones always show insanely high capacitances.
Eva, good point re: die size. Having come from the semicon industry, that should have occured to me! Thanks!
K-amps, I agree, there's nothing probably better than cracking the case open, but I'm afraid you can't do that without buying at least one, and the supplies here are too inconsistent for me to trust that if I get lucky getting a good one once, that it will always be 'good'...
Chris, thanks - as always, the voice of experience!
Well, let's get it on - who's going to design a nice, simple means of measuring capacitance?
- works on batteries
- interfaced to a uC
- measures in the low pF range...
- low voltage so that it doesn't bias the transistor's junction on...
Cheers!
K-amps, I agree, there's nothing probably better than cracking the case open, but I'm afraid you can't do that without buying at least one, and the supplies here are too inconsistent for me to trust that if I get lucky getting a good one once, that it will always be 'good'...
Chris, thanks - as always, the voice of experience!
Well, let's get it on - who's going to design a nice, simple means of measuring capacitance?
- works on batteries
- interfaced to a uC
- measures in the low pF range...
- low voltage so that it doesn't bias the transistor's junction on...
Cheers!
clem_o said:
The range for B-E junction capacitances in audio power transistors is more like 1 to 20 nanofarads. For C-B capacitances take one tenth of that.
p.s. I own a LCR meter from Monacor electronics (model LCR-4000) that does the job quite well. I think that most multimeters would be suitable as long as they employ small voltages.
Clem
Looks like you have a portable solution, without the use of the uC.
A capacitance meter implemented with a PIC would have been less accurate than the dedicated unit anyways.
Question for anyone who has cracked open flat packs (TO220, 247, 264):
Is the top face of the die in direct contact with the epoxy molding material, or is there a small space of air above it?
Although epoxy is not an outstanding conductor of heat, it's certainly better than air. I would hope a manufacturer would not have missed the opportunity to conduct heat away fro the die on both sides.
Looks like you have a portable solution, without the use of the uC.
A capacitance meter implemented with a PIC would have been less accurate than the dedicated unit anyways.
Question for anyone who has cracked open flat packs (TO220, 247, 264):
Is the top face of the die in direct contact with the epoxy molding material, or is there a small space of air above it?
Although epoxy is not an outstanding conductor of heat, it's certainly better than air. I would hope a manufacturer would not have missed the opportunity to conduct heat away fro the die on both sides.
Off Topic: TO3 vs flat device robustness
I've actually seen on a couple of occasions TO-3 transistors in which the glass seals failed.
In one case, whenever the transistor heated up during operation, some sizzling and bubbling could be seen/heard at the interface of the emitter pin where it meets the glass seal.
In the other instance, one of the pins started to rotate slightly, as the wire soldered to it was pulled around during heat sink installation. The amount of rotation soon increased to half a revolution.
In both above situations, the transistor had been handled extensively ( though not harshly) during prototyping and testing.
So it seems that although TO3 cases appear more robust, they are actually more susceptible to breach if overly handled. They are really meant to be taken out of the package, put in an application, and left alone.
The cylindrical pins lodged in a glass seal are a fragile construct. Any bending or torsional force can crack the glass or dislodge the pin. Compare that to the flat pins of the TO246, which cannot rotate, and have a greater surface area of contact with the embedding medium (epoxy) so better grip. Also epoxy has greater modulus of elasticity than glass, so it can flex slightly with the pin and not lose its grip.
Altough I always revered the TO3 because if it's superior thermal properties, there's no denying that it's geometry was inspired by the vaccuum tube mounting model: business end above the chassis, screw-mounted like a tube socket, leads sticking down below chassis, (where passive parts reside), cylindrical pins traversing a glass medium, leading to an airtight chamber inside where the business occurs. In a time (late 1950's) when electronic design semed like it would employ the chassis construction forever, this transistor was designed like a vaccuum tube.
For all intents and purposes, it's base/emitter pins have a robustness that's on par with the vaccuum tube.
Although I've heard it mentined that it takes 2-3 flatpack transistors to replace a TO-3, it seems that you need to be more careful in handling the TO3 than flat package devices.
Andy
I've actually seen on a couple of occasions TO-3 transistors in which the glass seals failed.
In one case, whenever the transistor heated up during operation, some sizzling and bubbling could be seen/heard at the interface of the emitter pin where it meets the glass seal.
In the other instance, one of the pins started to rotate slightly, as the wire soldered to it was pulled around during heat sink installation. The amount of rotation soon increased to half a revolution.
In both above situations, the transistor had been handled extensively ( though not harshly) during prototyping and testing.
So it seems that although TO3 cases appear more robust, they are actually more susceptible to breach if overly handled. They are really meant to be taken out of the package, put in an application, and left alone.
The cylindrical pins lodged in a glass seal are a fragile construct. Any bending or torsional force can crack the glass or dislodge the pin. Compare that to the flat pins of the TO246, which cannot rotate, and have a greater surface area of contact with the embedding medium (epoxy) so better grip. Also epoxy has greater modulus of elasticity than glass, so it can flex slightly with the pin and not lose its grip.
Altough I always revered the TO3 because if it's superior thermal properties, there's no denying that it's geometry was inspired by the vaccuum tube mounting model: business end above the chassis, screw-mounted like a tube socket, leads sticking down below chassis, (where passive parts reside), cylindrical pins traversing a glass medium, leading to an airtight chamber inside where the business occurs. In a time (late 1950's) when electronic design semed like it would employ the chassis construction forever, this transistor was designed like a vaccuum tube.
For all intents and purposes, it's base/emitter pins have a robustness that's on par with the vaccuum tube.
Although I've heard it mentined that it takes 2-3 flatpack transistors to replace a TO-3, it seems that you need to be more careful in handling the TO3 than flat package devices.
Andy
Hi Andy,
In over 35 years handling these transistors, I have never seen that happen. Many, many hundred TO-3 parts have passed through my hands. I'm not saying it can't happen, just I haven't seen it.
I have seen the pins burned off, and burned straight up into the case. At least they don't turn into shrapnel when they fail.
-Chris
In over 35 years handling these transistors, I have never seen that happen. Many, many hundred TO-3 parts have passed through my hands. I'm not saying it can't happen, just I haven't seen it.
I have seen the pins burned off, and burned straight up into the case. At least they don't turn into shrapnel when they fail.
-Chris
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.