14971	2018-08-02 12:19:49	Scott McDonald	Sig Gen Woes
Message	Date	From	Subject
Google has failed me, so I'm hoping this isn't too OT for this group. After doing a somewhat rigorous check of my trusty old, 300 buck flea bay, not calibrated this century I expect HP 8657A, freq looks good but it's output level looks high by about 20 -25 dBm. The decade to decade behavior seems about right thru the mid-range (but 20-25dB high), but the highest three decades are pretty much maxed out at the same level at about 20 dBm, which is consistent with the max output listed in the manual. My question, as best I can put it, is should I invest time trying to get the levels somewhere near accuracy? This is an unfamiliar beast, and I can't seem to find anything that discusses how these things age and fail. If it looks like a difficult to diagnose component-related failure situation, it's probably better for me to just measure the levels with a scope when using it, or max it out and run an attenuator, but if it sounds like something I might improve with just a scope and the manual, I'd be inclined to give it a go as it is (was) a nice piece of kit. My primary use of the generator has been aligning vintage gear, where level hasn't been much of an issue, but as I'm building HF receivers and trying to learn more about performance measurement, this is a pretty unsat situation, and the question is whether to invest time in this, or purpose build a few of the oscillators in EMRFD and call it a day. Any thoughts or other places to try are really appreciated. thanks, Scott ka9p
14972	2018-08-02 12:39:23	Steve65	Re: Sig Gen Woes
Hi Scott, You could try the HP/Agilent Test Equipment yahoo group. Some guys there are familiar with that signal generator. Steve, K8JQ
14973	2018-08-02 13:01:29	Thomas S. Knutsen	Re: Sig Gen Woes
In my opinion, you should fix the generator. Its a good generator in its price range, altough the phase noise is a bit high. It was supported by Agilent untill 2002, so its not that old. I got a couple around that gets used when the weigth of the HP8662A's is too high. There are a lot of information in the HP-agilent yahoo group on these generators: https://groups.yahoo.com/neo/groups/hp_agilent_equipment/ The 8656A, B and 8657A and B are quite similar and its probably one of the most common generators around, so there is a lot of information. The service manual has all the test, and HP service manuals are quite excelent at providing fault finding information. KO4BB should have the relevant manual for your generator. Follow the tests and it should be obvious what has failed. The output being above what you have set is not a common fault in these generators. Amplitude being low, not high is the most common error on these generators, and are usualy caused by someone transmitting into it. How is the amplitude tracking? is the offset constant or is it dependent on what amplitude you have selected? 73 de Thomas LA3PNA. Den tor. 2. aug. 2018 kl. 21:40 skrev Scott McDonald ka9p@aol.com [emrfd] <emrfd@yahoogroups.com>: Google has failed me, so I'm hoping this isn't too OT for this group. After doing a somewhat rigorous check of my trusty old, 300 buck flea bay, not calibrated this century I expect HP 8657A, freq looks good but it's output level looks high by about 20 -25 dBm. The decade to decade behavior seems about right thru the mid-range (but 20-25dB high), but the highest three decades are pretty much maxed out at the same level at about 20 dBm, which is consistent with the max output listed in the manual. My question, as best I can put it, is should I invest time trying to get the levels somewhere near accuracy? This is an unfamiliar beast, and I can't seem to find anything that discusses how these things age and fail. If it looks like a difficult to diagnose component-related failure situation, it's probably better for me to just measure the levels with a scope when using it, or max it out and run an attenuator, but if it sounds like something I might improve with just a scope and the manual, I'd be inclined to give it a go as it is (was) a nice piece of kit. My primary use of the generator has been aligning vintage gear, where level hasn't been much of an issue, but as I'm building HF receivers and trying to learn more about performance measurement, this is a pretty unsat situation, and the question is whether to invest time in this, or purpose build a few of the oscillators in EMRFD and call it a day. Any thoughts or other places to try are really appreciated. thanks, Scott ka9p
14974	2018-08-02 13:38:43	Tayloe, Dan (Noki...	Re: Sig Gen Woes
In doing receiver testing with my 8640B signal generators, I have accidently tripped a key and transmitted once or twice into my signal generator. Fortunately with QRP and a brief transmission, I did no damage. However, if this had been a 100w rig, I can see where I could have fried the attenuators in the signal generator. It seems to me likely that the step attenuator resistors in the first 2 attenuation steps are fried and need to be replaced. At the very least it seems to be a good place to start. Dan
14975	2018-08-02 16:44:40	Scott McDonald	Re: Sig Gen Woes
Great Steve, wasn’t aware of that, if Dan’s idea isn’t it, will head there next. 73 Scott Sent from my iPad
14976	2018-08-02 16:45:12	Scott McDonald	Re: Sig Gen Woes
Thanks Dan, will do, it would be great if that was it. Best, Scott Sent from my iPad
14977	2018-08-02 17:10:46	augustinetez	Re: Sig Gen Woes
Not necessarily the cause of your problem but there is mention on the Antique Radios forum (from 2008) of someone else with a similar problem. In his case, it was traced to a faulty reverse power detection module. (Google search term used was - HP 8657A attenuator fault) Terry VK5TM
14978	2018-08-02 19:00:18	lmarion	Re: Sig Gen Woes
That is an excellent place to start on any 8640. Long long ago in my first career in NIST Electronics labs, the several hundred 8640s I repaired and calibrated over 20 years, 90% of the time it was a bad attenuator. Leroy AB7CE
14979	2018-08-02 19:00:50	lmarion	Re: Sig Gen Woes
The 8657A has a reverse power protect circuit that tries to take the heat. Common problem.
14980	2018-08-03 08:45:08	Scott McDonald	Re: Sig Gen Woes
Thanks Terry, will take a look. Yeah, my searching could have been better if I had a clue :) It's fairly embarrassing how little I know about how that thing works, but I'm starting thru the manual now, definitely rooting for a bad attenuator :) cheers, Scott -----Original Message-----
14981	2018-08-03 08:46:16	Scott McDonald	Re: Sig Gen Woes
Thanks Thomas. The amplitude changes by the right amount in the middle of the range, I haven't seen any amplitude dependence - I can't measure the lowest few decades with my scope, and as I mentioned it's maxed out in the top two decades. Joined the HP group this PM and have been digging thru there as well. You guys are the best. I remember about 2 years ago when I started building stuff semi-seriously a friend telling me I'd probably end up spending more time with test equipment than radios, I'm starting to see what he meant :) Scott -----Original Message-----