HP 8901A is not detecting Frequency or RF level. - UK Vintage Radio Repair and Restoration Discussion Forum

stahodad · 25th Jun 2019, 3:14 am

Hello everyone,

I am working on a HP 8901A Modulation Analyzer that powers on ok. and seems to accept commands fine, but when a proper Frequency and level is applied the display goes from the two horizontal dashes to four dashes, but does not seem to lock onto the frequency.

Normally the display goes from two dashes to four momentarily and then displays the frequency. ( I call it the drum roll, please moment) .

It also has very low and constantly changing RF level display (that is not accurate at all), when that function is selected. There is a service keypad command for voltage level, and those seem to behave in a similar fashion, levels jumping around a few times a second and nowhere near the real value.

So far I have mostly looked at the RF input board (A15). Initially I had only the four dashes but managed to get the two dashes working when I swapped out the A17 Input mixer board from a spare parts 8901A. The parts analyzer actually has similar issues as well. It has me thinking the previous owner applied too much power to the input.

Any help with this analyzer would be much appreciated. It has quite a lot of on board diagnostics and special functions that are almost too numerous to mention, but I have to start somewhere.

I do have the downloadable service manual from Keysight. It is only a partial manual with some scans of diagrams that are hard to read, but it does have some useful information.

Thank you, Chris.

Radio Wrangler · 25th Jun 2019, 3:35 am

There are two ways of using this box. You can let it try to find an incoming frequency with its scanning algorithm, or you can tell it the frequency to sit and stare at. The dashes are what it shows during searching and level autoranging.

If there is a fault, the search isn't going to work, so you have to manually enter a frequency to make any progress.

The 8901 is a single-superhet down-converting to a low IF, with absolutely no image protection filtering. I can't remember the IF frequency, but the synth runs offset from the entered frequency.

The fault finding process is to enter a frequency, then probe to check the synthesiser is running and on the right frequency. Then you put in a signal and follow it down the receiver strip.

There's an HP journal article on the 8901 which you might find helpful.

The worst outcome would be if it's a fault in the frac-N synthesiser loop. Those things are complicated and use several irreplaceable custom parts. I've fixed them before, but I used to design them. Guiding someone to work their way through one would be a very large undertaking, but they seem fairly reliable, and there are fault finding flowcharts in the manual.

Anyway... type in frequency and see what happens. Check synth freq, and go from there

David

stahodad · 26th Jun 2019, 2:26 am

Hello David. Thank you for your reply and help with this analyzer.

Today I applied 200 Mhz signal at about 250 millivolts Pk to Pk. I hooked up CH1 of the oscilloscope to the output of the RF input board, A15, and CH2 is monitoring the applied signal level.

The CH1 signal was dancing around until I entered 200 Mhz on the 8901A keypad and the signal settled down to the first picture of the scope.
The frequency is about 199.8 Mhz, and about 88 millivolts pk to pk. ( I have not yet figured out if this is the normal output level). When I apply more power the levels on the scope increase as well. I did not go much higher than 500 millivolts Pk to Pk and this CH1 level went up to 120 millivolts of so.

Then I measured the output of the Input Mixer and that is what you see in the second and third picture of the scope. I might have to enter a command for this, because the 120 Hz output did not change when I increased the level or if I removed the 200 Mhz signal entirely. The last scope picture is of a different IF Mixer board I had, and the wave is a little different in shape. The red cable, OPT 3, did not have any real output.

I decided to work with the "parts" 8901 first and try to learn as much as I can, and then compare readings with the "good" 8901. I will try to find the 8901 bulletin you mentioned. I realize I may not get very far with this repair, but I hope to learn a few things along the way. Many Thanks, Cheers Chris.

Radio Wrangler · 26th Jun 2019, 4:54 am

Here it is:

https://www.hpl.hp.com/hpjournal/pdf...Fs/1979-11.pdf

I see it doesn't use Analog Phase Interpolation type Frac-N synthesis, it was just a bit too early for that. That makes things simpler.

Too much input will probably damage the mixer before the input attenuator burns out.

You need to enter, say, 200MHz on the keypad and get a counter on the LO signal.

David

IanG3XYV · 26th Jun 2019, 10:12 pm

Chris,

I am fortunate to have two working 8901As. They are superb examples of RF test equipment built in the best HP style of that era. Not only do they measure modulation levels but also RF power and signal frequency. The HP Journal article pointed to by David on the design philosophy is well worth reading.

As both of mine are working, I have not had to open them up to fault find, but they are designed for maintenance and repair with as ever an excellent service manual. I can recommend you following that to fault find. I have an original manual with full circuit diagrams. You say you have the partial Keysight download. If there are any essential bits missing, I should be able to scan for you.

There is also a very good HP/Agilent/Keysight forum here with lots of experts:

https://groups.io/g/HP-Agilent-Keysight-equipment

They should be able to help. There is lots of information there.

Good luck!

Ian

Radio Wrangler · 26th Jun 2019, 10:35 pm

One thing to be beware of with the 8901.... it leaks a significant amount of LO power OUT of the front panel input connector! Don't split the RF from a transmitter into the 8901 and a spectrum analyser - you'll think there is a major fault in the transmitter and it's crap from the 8901.

And a second thing to be beware of, the RF power measurement accuracy is rough, very rough on the 8901A. The later 8901B added a conventional power meter head (like the 8481 series for the 435 power meters)

stahodad · 28th Jun 2019, 2:08 am

Gentlemen, Thank you for the replies and the links for the 8901. I hope to have an update after the weekend or a few days later. Have a good weekend, Cheers Chris.

25th Jun 2019, 3:14 am	#1
stahodad Retired Dormant Member Join Date: May 2018 Location: Halifax, Nova Scotia, Canada. Posts: 11	HP 8901A is not detecting Frequency or RF level. Hello everyone, I am working on a HP 8901A Modulation Analyzer that powers on ok. and seems to accept commands fine, but when a proper Frequency and level is applied the display goes from the two horizontal dashes to four dashes, but does not seem to lock onto the frequency. Normally the display goes from two dashes to four momentarily and then displays the frequency. ( I call it the drum roll, please moment) . It also has very low and constantly changing RF level display (that is not accurate at all), when that function is selected. There is a service keypad command for voltage level, and those seem to behave in a similar fashion, levels jumping around a few times a second and nowhere near the real value. So far I have mostly looked at the RF input board (A15). Initially I had only the four dashes but managed to get the two dashes working when I swapped out the A17 Input mixer board from a spare parts 8901A. The parts analyzer actually has similar issues as well. It has me thinking the previous owner applied too much power to the input. Any help with this analyzer would be much appreciated. It has quite a lot of on board diagnostics and special functions that are almost too numerous to mention, but I have to start somewhere. I do have the downloadable service manual from Keysight. It is only a partial manual with some scans of diagrams that are hard to read, but it does have some useful information. Thank you, Chris. Attached Thumbnails

25th Jun 2019, 3:35 am	#2
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,867	Re: HP 8901A is not detecting Frequency or RF level. There are two ways of using this box. You can let it try to find an incoming frequency with its scanning algorithm, or you can tell it the frequency to sit and stare at. The dashes are what it shows during searching and level autoranging. If there is a fault, the search isn't going to work, so you have to manually enter a frequency to make any progress. The 8901 is a single-superhet down-converting to a low IF, with absolutely no image protection filtering. I can't remember the IF frequency, but the synth runs offset from the entered frequency. The fault finding process is to enter a frequency, then probe to check the synthesiser is running and on the right frequency. Then you put in a signal and follow it down the receiver strip. There's an HP journal article on the 8901 which you might find helpful. The worst outcome would be if it's a fault in the frac-N synthesiser loop. Those things are complicated and use several irreplaceable custom parts. I've fixed them before, but I used to design them. Guiding someone to work their way through one would be a very large undertaking, but they seem fairly reliable, and there are fault finding flowcharts in the manual. Anyway... type in frequency and see what happens. Check synth freq, and go from there David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

26th Jun 2019, 2:26 am	#3
stahodad Retired Dormant Member Join Date: May 2018 Location: Halifax, Nova Scotia, Canada. Posts: 11	Re: HP 8901A is not detecting Frequency or RF level. Hello David. Thank you for your reply and help with this analyzer. Today I applied 200 Mhz signal at about 250 millivolts Pk to Pk. I hooked up CH1 of the oscilloscope to the output of the RF input board, A15, and CH2 is monitoring the applied signal level. The CH1 signal was dancing around until I entered 200 Mhz on the 8901A keypad and the signal settled down to the first picture of the scope. The frequency is about 199.8 Mhz, and about 88 millivolts pk to pk. ( I have not yet figured out if this is the normal output level). When I apply more power the levels on the scope increase as well. I did not go much higher than 500 millivolts Pk to Pk and this CH1 level went up to 120 millivolts of so. Then I measured the output of the Input Mixer and that is what you see in the second and third picture of the scope. I might have to enter a command for this, because the 120 Hz output did not change when I increased the level or if I removed the 200 Mhz signal entirely. The last scope picture is of a different IF Mixer board I had, and the wave is a little different in shape. The red cable, OPT 3, did not have any real output. I decided to work with the "parts" 8901 first and try to learn as much as I can, and then compare readings with the "good" 8901. I will try to find the 8901 bulletin you mentioned. I realize I may not get very far with this repair, but I hope to learn a few things along the way. Many Thanks, Cheers Chris. Attached Thumbnails

26th Jun 2019, 4:54 am	#4
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,867	Re: HP 8901A is not detecting Frequency or RF level. Here it is: https://www.hpl.hp.com/hpjournal/pdf...Fs/1979-11.pdf I see it doesn't use Analog Phase Interpolation type Frac-N synthesis, it was just a bit too early for that. That makes things simpler. Too much input will probably damage the mixer before the input attenuator burns out. You need to enter, say, 200MHz on the keypad and get a counter on the LO signal. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

26th Jun 2019, 10:12 pm	#5
IanG3XYV Pentode Join Date: Sep 2014 Location: Bedford, Bedfordshire, UK. Posts: 174	Re: HP 8901A is not detecting Frequency or RF level. Chris, I am fortunate to have two working 8901As. They are superb examples of RF test equipment built in the best HP style of that era. Not only do they measure modulation levels but also RF power and signal frequency. The HP Journal article pointed to by David on the design philosophy is well worth reading. As both of mine are working, I have not had to open them up to fault find, but they are designed for maintenance and repair with as ever an excellent service manual. I can recommend you following that to fault find. I have an original manual with full circuit diagrams. You say you have the partial Keysight download. If there are any essential bits missing, I should be able to scan for you. There is also a very good HP/Agilent/Keysight forum here with lots of experts: https://groups.io/g/HP-Agilent-Keysight-equipment They should be able to help. There is lots of information there. Good luck! Ian

26th Jun 2019, 10:35 pm	#6
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,867	Re: HP 8901A is not detecting Frequency or RF level. One thing to be beware of with the 8901.... it leaks a significant amount of LO power OUT of the front panel input connector! Don't split the RF from a transmitter into the 8901 and a spectrum analyser - you'll think there is a major fault in the transmitter and it's crap from the 8901. And a second thing to be beware of, the RF power measurement accuracy is rough, very rough on the 8901A. The later 8901B added a conventional power meter head (like the 8481 series for the 435 power meters) __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

28th Jun 2019, 2:08 am	#7
stahodad Retired Dormant Member Join Date: May 2018 Location: Halifax, Nova Scotia, Canada. Posts: 11	Re: HP 8901A is not detecting Frequency or RF level. Gentlemen, Thank you for the replies and the links for the 8901. I hope to have an update after the weekend or a few days later. Have a good weekend, Cheers Chris.