Path: ...!eternal-september.org!feeder2.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Cursitor Doom <cd999666@notformail.com>
Newsgroups: sci.electronics.design
Subject: Re: Curve Tracers
Date: Thu, 21 Nov 2024 23:35:13 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 124
Message-ID: <vhog3g$rlc2$1@dont-email.me>
References: <vhkhcg$2ip8$2@dont-email.me>
	<o50sjjpmjmrdbvs9po8inhcdn961a0923g@4ax.com> <vhl842$70jp$1@dont-email.me>
	<d5hsjj967gqpgmoja3ndgi5re9o38p3ki5@4ax.com> <vhlt6a$9hat$3@dont-email.me>
	<j10tjjt05qolld20qtpfpg2nmb1vhcpsrt@4ax.com> <vhn7gk$kdip$1@dont-email.me>
	<v6mujjtghinmo0gf279taqv4osv63b7ir2@4ax.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Date: Fri, 22 Nov 2024 00:35:13 +0100 (CET)
Injection-Info: dont-email.me; posting-host="877b91f196afc527683bd58c23dff4b6";
	logging-data="906626"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+edsOOz7cdeparlcuNq+m0YGHRTz4w20Q="
User-Agent: Pan/0.149 (Bellevue; 4c157ba)
Cancel-Lock: sha1:iciHOIwdBahI0LdxW/LPBZNrov8=
Bytes: 6468

On Thu, 21 Nov 2024 08:03:00 -0800, john larkin wrote:

> On Thu, 21 Nov 2024 12:02:28 -0000 (UTC), Cursitor Doom
> <cd999666@notformail.com> wrote:
> 
>>On Wed, 20 Nov 2024 16:37:39 -0800, john larkin wrote:
>>
>>> On Thu, 21 Nov 2024 00:00:10 -0000 (UTC), Cursitor Doom
>>> <cd999666@notformail.com> wrote:
>>> 
>>>>On Wed, 20 Nov 2024 12:27:04 -0800, john larkin wrote:
>>>>
>>>>> On Wed, 20 Nov 2024 18:00:34 -0000 (UTC), Cursitor Doom
>>>>> <cd999666@notformail.com> wrote:
>>>>> 
>>>>>>On Wed, 20 Nov 2024 07:40:08 -0800, john larkin wrote:
>>>>>>
>>>>>>> On Wed, 20 Nov 2024 11:32:32 -0000 (UTC), Cursitor Doom
>>>>>>> <cd999666@notformail.com> wrote:
>>>>>>> 
>>>>>>>>Gentlemen,
>>>>>>>>
>>>>>>>>Curve tracers reveal useful info about the dynamic characteristics
>>>>>>>>of semiconductors and make designing for same much more
>>>>>>>>predictable and dependable than relying on spice models and
>>>>>>>>simulation alone. But they're typically rare beasts and expensive
>>>>>>>>to come by and boat anchor varieties are seriously heavy and
>>>>>>>>bulky.
>>>>>>>> I think therefore that a curve tracer would make an excellent
>>>>>>>> project,
>>>>>>>>using the X&Y inputs of a scope as the display. Has anyone here
>>>>>>>>attempted this? I'd be interested to know what the main challenges
>>>>>>>>are likely to be.
>>>>>>>>
>>>>>>>>-CD
>>>>>>> 
>>>>>>> I've considered it. It would be much more valuable if it traced
>>>>>>> capacitances too.
>>>>>>
>>>>>>A VNA is best for that. I've been using my big HP one this afternoon
>>>>>>to label some random bunches of caps and inductors whose values I
>>>>>>couldn't read. Trouble is, there doesn't seem to be much agreement
>>>>>>on what frequency to test these devices at. I've heard 100khz, 1Mhz,
>>>>>>10Mhz and 100Mhz mentioned from different sources. I'm pretty sure
>>>>>>the 100Mhz testing is for RF specified devices only, though. From
>>>>>>what I've seen, there's no reliable alternative to testing each
>>>>>>batch for oneself, because (certainly with ancient NOS leaded (as in
>>>>>>non-SMD) stock) you have no idea what frequency the factory tested
>>>>>>them at and different manufacturers in different countries at
>>>>>>different times used different methods!
>>>>> 
>>>>> What I want is C-V curves. I guess a VNA can do that with a some
>>>>> bias tees and various power supplies.
>>>>
>>>>If you only want C/V curves, what's wrong with a plain 'scope and a
>>>>pulse generator?
>>> 
>>> How would you do that?
>>
>>Ah! Well, I obviously mis-read your earlier statement. My bad, soz.
>>
>>>>>>> The display should be on a computer and the data archived.
>>>>>>
>>>>>>For people such as yourself, certainly. For me as a hobbyist, a CRT
>>>>>>is fine.
>>>>> 
>>>>> You have to read the analog screen and write down numbers. Or take a
>>>>> picture.
>>>>
>>>>It would only be a (slight) issue if I needed to share an image with a
>>>>third party. Aside from that, there's nothing I couldn't live with.
>>>>You're obviously requiring more than that, though.
>>>>
>>>>>>> I often just set up a breadboard and test parts. Last week I blew
>>>>>>> out some power mosfet gates. Some have internal zeners; this one
>>>>>>> doesn't.
>>>>>>
>>>>>>I can kill a MOSFET at 50' just by looking at it (certainly in
>>>>>>winter). In fact I shock myself from all the static I build up and
>>>>>>it ain't funny. Having dry skin is great for when I touch HV by
>>>>>>accident (happens quite a lot) but the flip side is I accumulate and
>>>>>>hold static charge like no one else I know.
>>>>> 
>>>>> Some fets have protective gate zeners. They typically clamp at +-40
>>>>> volts.
>>>>> 
>>>>> Here in San Francisco, we never get static zapped. High humidity.
>>>>
>>>>So you never need to wear a wrist/earth strap? That's a big plus. I
>>>>find them *so* restrictive and irritating when they (invariably) catch
>>>>on something.
>>>>
>>>>>>> Most mosfets drain avalanche, but the voltage is never specified,
>>>>>>> in fact deliberately hidden. I have to measure that.  A good curve
>>>>>>> tracer should measure that too.
>>>>>>> 
>>>>>>> Currents should go down to picoamps.
>>>>>>> 
>>>>>>> The fact that there aren't many curve tracers for sale suggests a
>>>>>>> small market.
>>>>>>
>>>>>>But a market, though small perhaps, there must be.
>>>>> 
>>>>> Sure. It would be a fun project.
>>>>
>>>>And with thanks to Trump's tariffs, you might even make a few buxx out
>>>>of it, too.
>>> 
>>> Is there going to be tariffs on USB capacitance-measuring curve
>>> tracers?
>>
>>Not specifically, but I imagine Trump will want tariffs on Chinese
>>imports in general, which should give some breathing space profit-wise
>>to US manufacturers.
>>I have to say I do admire Trump's job creation ideas, which should
>>produce solid results, as opposed to old Joe's approach of destroying US
>>jobs, seemingly on purpose. I say old Joe's approach, but what I really
>>mean is the approach of the Globalists who tell him and his ilk what to
>>do.
> 
> We should stop paying the Chinese for the shipping costs of cheap junk,
> for starters.

The US does that too?? Wow. I thought that was just a European thing.