Warning: mysqli::__construct(): (HY000/1203): User howardkn already has more than 'max_user_connections' active connections in D:\Inetpub\vhosts\howardknight.net\al.howardknight.net\includes\artfuncs.php on line 21
Failed to connect to MySQL: (1203) User howardkn already has more than 'max_user_connections' active connectionsPath: ...!eternal-september.org!feeder2.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Cursitor Doom Newsgroups: sci.electronics.design Subject: Re: Curve Tracers Date: Fri, 22 Nov 2024 18:11:04 -0000 (UTC) Organization: A noiseless patient Spider Lines: 147 Message-ID: References: <09hvjjta6p3t9ffe2ji2j30hnautcvdmqa@4ax.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Fri, 22 Nov 2024 19:11:04 +0100 (CET) Injection-Info: dont-email.me; posting-host="877b91f196afc527683bd58c23dff4b6"; logging-data="1354769"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX195QCVRuv7yHJI0FtizE6F/eir91TtlBj8=" User-Agent: Pan/0.149 (Bellevue; 4c157ba) Cancel-Lock: sha1:rtcoizQrFjPNKxen43ICAA3ovXY= Bytes: 7692 On Thu, 21 Nov 2024 15:43:21 -0800, john larkin wrote: > On Thu, 21 Nov 2024 23:35:13 -0000 (UTC), Cursitor Doom > wrote: > >>On Thu, 21 Nov 2024 08:03:00 -0800, john larkin wrote: >> >>> On Thu, 21 Nov 2024 12:02:28 -0000 (UTC), Cursitor Doom >>> 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 >>>>> 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 >>>>>>> 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 >>>>>>>>> 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. > > The Chinese are a "disadvantaged country" per the International Postal > Union. That needs updating! > They will ship a cheap ebay thing here for free. It would cost me a > bundle to ship it back. Same here. Function generator arrived today direct from China and I didn't have to pay a penny in sales tax for it. No duties whatsoever. Express postage was free, too. The treasuries of Europe must be losing untold tens of billions in taxes due to this. And yet everyone has to pay sales tax on domestically-produced goods! It's madness on steroids. They'll probably claim it would cost more to collect the tax than the revenue doing so would generate, but that's indefensible BS plain and simple.