Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail From: Cursitor Doom Newsgroups: sci.electronics.design Subject: Re: Distorted Sine Wave Date: Fri, 31 May 2024 17:29:47 -0000 (UTC) Organization: A noiseless patient Spider Lines: 111 Message-ID: References: <3lcf5jd7li0a3c0fgddt7o8lnfocvls2pr@4ax.com> <48bd78e1-7da8-3bba-2879-d22962203fa3@electrooptical.net> <9olh5j9al34fhrebr4grqq8h6c8javjpp1@4ax.com> <1n0i5jh257hiinlj2dhaatlo11s33m5n0e@4ax.com> <9k2i5jpfhu3ncfpm28ukusrok4hugal80s@4ax.com> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Injection-Date: Fri, 31 May 2024 19:29:47 +0200 (CEST) Injection-Info: dont-email.me; posting-host="5298e01e4585ab4377a4cf319f61075c"; logging-data="2448975"; mail-complaints-to="abuse@eternal-september.org"; posting-account="U2FsdGVkX1+JrMCKkJnVJtFKtWslUrmtbqEFbwdg8dw=" User-Agent: Pan/0.149 (Bellevue; 4c157ba) Cancel-Lock: sha1:jkxfxt7Wb6iMXWboaNYRPbNCNQw= Bytes: 6128 On Fri, 31 May 2024 11:34:46 -0400, Joe Gwinn wrote: > On Thu, 30 May 2024 16:24:25 -0700, john larkin wrote: > >>On Thu, 30 May 2024 18:53:29 -0400, Joe Gwinn >>wrote: >> >>>On Thu, 30 May 2024 14:56:30 -0700, john larkin wrote: >>> >>>>On Thu, 30 May 2024 16:06:47 -0400, Joe Gwinn >>>>wrote: >>>> >>>>>On Thu, 30 May 2024 15:02:44 -0400, Phil Hobbs >>>>> wrote: >>>>> >>>>>>On 2024-05-29 18:59, Joe Gwinn wrote: >>>>>>> On Wed, 29 May 2024 22:11:47 -0000 (UTC), Phil Hobbs >>>>>>> wrote: >>>>>>> >>>>>>>> Cursitor Doom wrote: >>>>>>>>> On Wed, 29 May 2024 13:42:13 -0700, john larkin wrote: >>>>>>>>> >>>>>>>>>> On Wed, 29 May 2024 21:43:54 +0200, Arie de Muijnck >>>>>>>>>> >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>>> On 2024-05-29 19:07, Cursitor Doom wrote: >>>>>>>>>>>> Gentlemen, >>>>>>>>>>>> >>>>>>>>>>>> Whilst fault-finding on my HP 8566B spectrum analyzer, I've >>>>>>>>>>>> found the 10Mhz reference oscillator is generating an >>>>>>>>>>>> 'unsatisfactory waveform' >>>>>>>>>>>> which may be causing the device to be unable to lock it's >>>>>>>>>>>> main PLL. I've come across this waveshape before, but mostly >>>>>>>>>>>> with oscillators I was building and in the process of trying >>>>>>>>>>>> to iron out the wrinkles of and certainly NOT a critical >>>>>>>>>>>> reference oscillator from a respected manufacturer. Can >>>>>>>>>>>> anyone tell what's most likely going on here? >>>>>>>>>>>> >>>>>>>>>>>> https://disk.yandex.com/i/z6fYbeVfPRK7aA >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> Looks like reflections in the cable. Try the 50 Ohm >>>>>>>>>>> termination. >>>>>>>>>>> >>>>>>>>>>> Arie >>>>>>>>>> >>>>>>>>>> If the drive is a sine wave, a cable can't generate that 2nd >>>>>>>>>> harmonic. >>>>>>>>> >>>>>>>>> I don't understand how a reflection can account for it either. >>>>>>>>> THe cable's only 4' long! However, with the 50 ohm input >>>>>>>>> enabled, the 2nd harmonic disappears. It's just one of those >>>>>>>>> inexplicable mysteries that no one knows the answer to. :) >>>>>>>> >>>>>>>> That’s pretty diagnostic. There must be an LC filter on the >>>>>>>> output—mis-terminating it will cause all sorts of >>>>>>>> frequency-response whoopdedoos. >>>>>>> >>>>>>> It also occurs to me that if there is a diode in series with a >>>>>>> resistor somewhere, the impedance presented to the feed coax may >>>>>>> be 50 ohms for positive input voltage, and say 10 Kohm for >>>>>>> negative. At the very least one could get an inverted reflection >>>>>>> on negative. >>>>>>> >>>>>>> Joe Gwinn >>>>>>> >>>>>>Yeah, or an emitter follower. Good point. >>>>> >>>>>Plus some LC filter wiggles, to distort and smooth things. >>>>> >>>>>I dug around and found a copy of the HP 8566B spectrum analyzer >>>>>service manual. The 10 MHz ref input is an amplifier driving a >>>>>mixer, >>>>>with not hint of for instance a TTL input. So, the problem must be >>>>>elsewhere. Or, it's just busted. >>>>> >>>>>. >>>>> >>>>>Joe Gwinn >>>> >>>>334 pages! Where is the issue? >>> >>>The pagination was unclear, so I didn't see a way to say. How I found >>>it was to look at the schematics. Look for module A22. Don't think >>>search works on such images, but visual search didn't take that long. >>> >>>Joe Gwinn >> >>Too much work for free consulting. > > I took another look. There is a page reference on the right edge, near > the bottom, which is not marked as a page X of Y, but is. > > Anyway, look at pages 59 and 77. At 77, look for A22 in the lower left > region. Just above, look for INT and EXT jacks. The EXT jack is where > an external reference enters. > > Joe Gwinn I don't believe the 10Mhz ref osc is the problem, John. The 2nd harmonic distortion goes away when the scope input impedance is set to 50 ohms. There is some slight distortion on it, but not enough to cause an out-of- lock error. Furthermore, since that osc is the pace-setter for every other module in every other loop in this analyzer, its failure would give rise to way more error messages than a mere "YTO unlock" as it stands at present. The manual suggests the most likely areas where the fault is located are in one of the boards A19, A20, A21 or A11. If A22 were the culprit, there would be over a dozen error messages.