Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: john larkin <JL@gct.com>
Newsgroups: sci.electronics.design
Subject: Re: Datasheet-flation?
Date: Thu, 28 Nov 2024 19:15:26 -0800
Organization: A noiseless patient Spider
Lines: 62
Message-ID: <3qbikjdvcnp41t8vnqtggo5nidqb9r654f@4ax.com>
References: <7qfi1l-0kq11.ln1@coop.radagast.org> <viatur$o8vt$1@news.eternal-september.org> <em5ikjded88fch9le9njlvsb8mcins9dpt@4ax.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Injection-Date: Fri, 29 Nov 2024 04:15:29 +0100 (CET)
Injection-Info: dont-email.me; posting-host="91f6c56077a3bd8c399c29b307a15535";
	logging-data="988469"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1/8I6ZsWbeDaPBfFvqioaH6"
User-Agent: ForteAgent/8.00.32.1272
Cancel-Lock: sha1:ZydeBefy5mwqnC7+stUFUND6bWw=
Bytes: 3560

On Thu, 28 Nov 2024 17:56:59 -0800, john larkin <JL@gct.com> wrote:

>On Thu, 28 Nov 2024 18:22:03 -0500, ehsjr <ehsjr@verizon.net> wrote:
>
>>On 11/26/2024 10:21 PM, Dave Platt wrote:
>>> Back on the original subject a bit: it's not just huge data sheets
>>> which can be... well, less than wonderful.  I invite you to take a
>>> look at the sheet for the Skyworks SKY65050-372LF low noise
>>> transistor.  The evaluation board schematic on page 8, and its
>>> associated bill of materials, invite careful examination.
>>> 
>>> The fact that they went to the trouble of actually _saying_ that
>>> the schematic diagram is a mess, rather than actually fixing it...
>>> well, I assume this was a Management decision.  I think I'd
>>> have been ashamed to be the Engineer, though.
>>> 
>>
>>It seems that the block diagram on page 1 may establish
>>the caliber of the documentation. :-(
>>
>>(What's the emoticon for hold your nose?)
>>Ed
>
>It's not even a block.
>
>The RF geeks like to call fet pins RF_IN and RF_OUT and RF_COM,
>whereas normal people say gate, drain, source. Maybe compromise on
>SOME and MORE and NONE.
>
>And their biasing algorithm is usually "turn the pot until it works."
>
>I use $100 RF switches and PHEMTs and MMICs and $300 distributed amp
>chips in time domain, to make pulses. The data sheets are pretty much
>useless to me. You never know if a part will actually work under 100
>MHz (some won't) , or if that's as low as their VNA will go. 
>
>Mini-Circuits has assured me that they have no interest in amplifying
>pulses and will never have Spice models for their parts. 
>
>I wonder if Q-spice is intended to bring RF design into the 20th
>Century. It's the 21st already so it's overdue.
>
>We had to add a DC block here, and it didn't help that some of the RF
>parts were poorly specified.
>
>https://www.dropbox.com/scl/fi/3j1eyfu850n5bb3xwlx10/DC_Block.jpg?rlkey=055ej3zuo5siukis8q0aaj8ft&raw=1
>

And for SOT-143 fets, i believe that every possible pin numbering
scheme has been used, including the FxxxR (reversed pinout) versions,
CW and CCW numbers.

And the bigfoot can be on the left or the right side. It's usually a
source pin. The one across from it is the gate or sometimes the drain.
Not to mention enhancement and depletion and wirebonds tuned to one
band.

So when CEL EOLs a part, which they enjoy doing, it's a nuisance to
find a replacement.

https://www.dropbox.com/scl/fo/5rpsajll41lkiojh8qg6t/ALjyqBI8R3BMI0CXQFpOZxs?rlkey=m6ijaefcsso42hwc4zwyjhcz6&dl=0