The PAD-1 picoamp diode was/is just a jfet with the source and drain connected. It's expensive, and has a lot of capacitance and series resistance. The BFT25 is a way superior diode. I recall that Ccb is a few tenths of a pF.
The problem with fA current measurements is that it takes forever for things to settle down, and any change in the electrostatic environment, like walking or breathing in the same room, wrecks things.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Put it in a shielded box. The cookie tin you use for phase noise measurements should do fine.
I do have a shield box to drop over my fA measurement thing. But it still takes ages to settle down.
Next time I do fAs, I'll use the capacitor charging idea. It will take time too.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Is the box open at the bottom? I think it needs to be closed all around.
There's the matter of insulators with surface-charge and dielectric-absorption issues.
--
Thanks,
- Win
Of course not. That wouldn't work.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Low noise transistors become noisy transistors after the b-e junction has been zenered a few times. I didn't have an opportunity to do a rigorous before-after gain comparison but it didn't seem to change dramatically.
-- Grizzly H.
Others have observed a serious loss in beta. There are rumors of ways to resurrect the poor beast, but if its role in life is as a low-capacitance zener clamp, who cares? When it's wired up on a PCB, with the base and collector shorted, there's little else that it could ever do!
--
Thanks,
- Win
Consider a UV LED inside the shield box; nothing discharges (leaks) better than a photoconductor (and almost all insulators ARE at least a little photoconductive).
A few milliseconds after the LED is turned off, the insulators will insulate again.
On a sunny day (22 Jun 2019 10:58:12 -0700) it happened Winfield Hill wrote in :
Sure,. but I was thinking the same thing did you try a zillion pulses and see if the curves are still the same after that? I mean something must physically change in those transistors, especially at high current?
Is that a plastic 8-pin dip socket?.. those can leak too.
Huh.. that's simple.. I think I did the somewhat the same thing, different opamp and the resistor on the bottom, to gnd.
Plastics should be mostly suspect in terms of leakage, I've had some issues with 8-pin dip plastic sockets. When there's no spec from the manufacturer. Then you learn to test every new order... some sales /rep guy said they will change the plastic mix for the same part. You end up ordering a bunch at one time.
My biggest R is a 1G.
George H.
Just don't put them back in the drawer after you finish playing.
-- Grizzly H.
One of them went into the trash can! The rest are taped to the page of data, as evidence.
--
Thanks,
- Win
On a sunny day (Sun, 23 Jun 2019 11:43:10 -0400) it happened mixed nuts wrote in :
Winfield makes an assumption, only testing can confirm it stays working as a zener in the long run. If anything physically changes in that transistor why should it not affect the Vbe zener. Not testing is bad practice.
Any experience on what is short enough wavelength for this ? I happen to have a rather large stash of 280nm LEDs..
-- mikko
Have you tried that? Got any numbers?
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Don't worry, my setup is test-ready. I'm using a $10k SMU that I got on eBay for $5k, and it has scripts for doing extended tests, measuring as you go. But I would also add a comment, this isn't my scheme, many have been using it for decades, and I suppose some did thorough tests. Also, events this is meant to protect against don't happen very often. For now, the 10A projection is a dotted line in the book, but it'll be tested before going to publication.
--
Thanks,
- Win
On a sunny day (23 Jun 2019 10:14:04 -0700) it happened Winfield Hill wrote in :
OK, well I am curious as to what you will find,.
5 k$ is a lot of money, something that generates 10A pulses should not be that expensive? Very long time ago boss asked me if I could design a transistor curve tracer. Took a day or 2, used relays too, and a scope. I am at a loss here:
cer.
Source Measure Unit, ~ a fancy powersupply
On Sunday, June 23, 2019 at 2:22:00 PM UTC-4, Lasse Langwadt Christensen wr ote:
racer.
Seems like an odd name for that. I assume it has some special properties b eyond a power supply which the name refers to?
I knew part of the name without looking it up, "unit"... lol
-- Rick C. - Get 1,000 miles of free Supercharging - Tesla referral code - https://ts.la/richard11209
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.