|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
20th Apr 2017, 4:27 pm | #1 |
Diode
Join Date: Apr 2017
Location: Anlaby Common, Hull, East Riding of Yorkshire
Posts: 4
|
HP 8569A Spectrum Analyser fault
I wonder if anyone can help me to pinpoint a problem on my HP 8569A analyser?
It shows up when the resolution is cranked down to 50kHz or less at any centre frequency as sprogs at +/- 32kHz, and multiples, from the carrier being looked at. Pease see attached screen shots, one showing the output of a 1152MHz local oscillator and the other is the 100MHz output from a HP8657B sig gen. I've used several carrier sources at many different frequencies and get the same result so I'm fairly confident it's an artefact of the analyser. Any thoughts would be much appreciated. Thanks Clive |
23rd Apr 2017, 5:37 pm | #2 |
Nonode
Join Date: Nov 2015
Location: Welwyn Garden City, Hertfordshire, UK.
Posts: 2,015
|
Re: HP 8569A Spectrum Analyser fault
Hello Clive, not many takers for this....
Has it any dc/dc converters? 32kHz sprogs are v.suspicious of a noisy one causing modulation somewhere. |
23rd Apr 2017, 7:25 pm | #3 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,902
|
Re: HP 8569A Spectrum Analyser fault
It should only have resolutions in a 1-3-10-30-100... series so I'm not sure about 50kHz resolution.
Unlock the resolution bandwidth knob from the resolution bandwidth knob and see if it changes with either. It's possible you always have sprogs and only certain spans and degrees of selectivity will show them up. The local oscillator is a YIG and is magnetically tuned. The analyser is quite sensitive to fields from things near it. There are tantalum electrolytics in the amplifiers which drive the YIG coils, including one with a load resistor I expect. If they suffer high ESR, the current driver stability can be ruined. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
23rd Apr 2017, 9:44 pm | #4 |
Nonode
Join Date: Nov 2015
Location: Welwyn Garden City, Hertfordshire, UK.
Posts: 2,015
|
Re: HP 8569A Spectrum Analyser fault
Hi David, a very complicated beast. From what I could see in the service info nothing likely except the EHT gen for the tube which runs at 40-45kHz.
Those spurs are not in that range so could be your theory about the YIG osc and pre-sel filter is much better. I wish I had one of these. I do have an HP8559 but it has inherited the disintegrated switch syndrome...the contacts have fallen out of the bottom and are all lost now. An HP8558 went the same way. It is such a shame! I keep it but I think it is beyond hope to fix it - but that is straying off topic. |
24th Apr 2017, 7:12 pm | #5 |
Diode
Join Date: Apr 2017
Location: Anlaby Common, Hull, East Riding of Yorkshire
Posts: 4
|
Re: HP 8569A Spectrum Analyser fault
Thanks for the responses Jon and David.
I can't see any DC-DC converters in the circuit and frankly I think the SA design too old for that type of device. The power supply is linear with all sorts of pos and neg rails so unless an isolated supply was needed I don't think I'm going to find one. The RB sequence on the 8569A is 1-2-5-10-20-50 as per the attached photo of the RB control knob which also shows traces of Araldite used to repair it twice. Seems to be a common problem. What I can't get my head around at the moment is why it only shows prominently when the RB is set to 50kHz or less. It is just there on the 100kHz setting but it's way down in the noise floor. There must be a clue in there somewhere but it eludes me. I too wondered whether there was some sort of ripple on the tuning current going to the YIG causing some FM of the oscillator. I suppose I could use the Rigol SA to check (that only goes up to 1.5GHz) by looking at the IF from mixing the local oscillator output and a 2GHz source, setting the Rigol into store mode and leaving it for a few minutes. Alternatively I could set to 8569A to zero sweep and look at the IF in real time but I'm not sure how much of the YIG tuning that disables. It should leave to DC paths intact. Does that make sense? Clive |