Path: ...!weretis.net!feeder9.news.weretis.net!news.quux.org!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 12:02:28 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 111
Message-ID: <vhn7gk$kdip$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>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Injection-Date: Thu, 21 Nov 2024 13:02:32 +0100 (CET)
Injection-Info: dont-email.me; posting-host="d622c9676f778f90c0fa53e3e54e2994";
	logging-data="669273"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/yLFdVyqdxUKGsRR/KRAFJQALWko4mw14="
User-Agent: Pan/0.149 (Bellevue; 4c157ba)
Cancel-Lock: sha1:VvYfSe5D+h29wss7HAgwMPXShkU=
Bytes: 5883

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.