Results 1 to 8 of 8

Thread: I bought service monitors

  1. #1
    Join Date
    Sep 24, 2012
    Posts
    123
    Thanks
    0
    Thanked 3 Times in 3 Posts

    Default I bought service monitors

    I think that the days of spectrum analysis by watching the s-meter and spinning the tuning knob, and deviation checking with a vfo and oscilloscope, are soon to be in my past.

    I went to the evil auction site with $400 and came out with two hp 8920ish communications analyzers. One of them has marker and sweep.

    One needs a firmware update and some cut cables restored, to get rid of the requirement for some external measurement box.

    The other one is a "it don't power on" mystery, it might be complete, maybe not. It has a power supply that is used in other instruments.

    Who knows what will become if either of them, and how many tests they will fail once brought into working configurations.

    One thing is for sure. Neither of them have hefty inputs, so for a while, my coupler remains this wire stub that picks up leakage from my hillbilly DIY dummy load.


  2. #2
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,235
    Thanked 1,531 Times in 738 Posts
    Country: Christmas Island

    Default

    Decent monitors, but as you noticed the power handling on them isn't the best, there are 60W (100?) versions but a lot are the 5W versions made to test cell phones. An external dummy load/attenuator is necessary but one of the more charming features of them is they will do LTR encode/decode. The sweep gen function is nice too for doing duplexers if you can get them running.

    Here's a couple of Operator's Manuals:
    Attached Files Attached Files

  3. The Following User Says Thank You to Alpha For This Useful Post:

    NFRad (Nov 12, 2014)

  4. #3
    Join Date
    Jan 19, 2013
    Location
    In Your Network
    Posts
    2,730
    Thanks
    3,063
    Thanked 1,965 Times in 992 Posts
    Country: Holy See

    Default

    There is a test equipment refurbisher in Las Vegas that will sell you the 100watt upgrade kits as well as other parts. He sells on eBay.

  5. #4
    Join Date
    Sep 24, 2012
    Posts
    123
    Thanks
    0
    Thanked 3 Times in 3 Posts

    Default

    Woo woo ! One of them arrived today. The lesser of the two: it's an 8922M.

    It's not easy to use. So far I've found the spec an, af gen, rf gen, and rf analyzer (fm deviation meter.)

    The internal attenuator is bypassed because it is configured for some external box. So far everything looks on point but the rf gen is off by about +13MHz. Either I need to make an adjustment, remember to dial it down, or find the firmware for this machine that doesn't expect the external box.

    There is no SSB/AM demod in this one. I should have an 8924c next week!

  6. #5
    Alpha's Avatar
    Alpha is offline T S - Moderator
    CS Forums $upporter
    Join Date
    Feb 12, 2012
    Location
    Directly above the center of the earth.
    Posts
    2,780
    Thanks
    1,235
    Thanked 1,531 Times in 738 Posts
    Country: Christmas Island

    Default

    It might be the crystal oven is dead. In other monitors I found when the heater goes out the monitor goes off by many Mhz like that. Try feeding it an external 10 Mhz reference and see if it's still off freq, that's the easiest way to tell if it's the reference standard oven or not.

  7. #6
    Join Date
    Sep 24, 2012
    Posts
    123
    Thanks
    0
    Thanked 3 Times in 3 Posts

    Default

    Cool! I took the lid off to make sure everything is ok. Some good things to see here, like that some of HP did not devolve into cheap plastic boxes for everything like they had done with their scope and small logic analyzer lines.

    I think the ocxo is ok. The spec an is on frequency and this morning I watched it "line up" as it got warm. I did have to calibrate the reference and the receiver and spec an are right on point, and it held overnight. I used an HF radio (2.5ppm) to get 10MHz to calibrate the ref.

    The machine is failing selftests because it tries to test the receiver using the generator which is off frequency. The generator being off might be because of "sum loop" or "upconverter" problems or miscalibration.

    I'm not going to try to mess with making sure the input and output levels are ok until I get the generator on frequency. I think that I can set the generator levels in dBm and use an oscilloscope to verify them below 250MHz (my scope is generally 500Mhz front end and 500MSa/S.)

    The self-calibration programs and other rom routines are written in 'HP Instrument BASIC' and I've downloaded them to look at. This is very interesting. The BASIC programs manipulate the machine as if it is a remote controller with a strange GPIB address. By reading the test programs I learn how to manually set up the machine to replicate these tests, either using the front panel or NI MAX. And there is BASIC programming shell on the serial port. I deleted a program statement which stopped the calibration program (because the 1800/1900 MHz box is not connected) and successfully ran some kind of level calibration. (yeahh using the 75 ohm pads I have laying around.) Now both input ports read the same levels.

    Regardless of what I do with it today it's going to sit powered on and "baking" to try and charge a memory card I have with dead batteries.

    This thing is a lot of fun.

  8. #7
    Join Date
    Sep 24, 2012
    Posts
    123
    Thanks
    0
    Thanked 3 Times in 3 Posts

    Default

    I found a replacement module and fixed the 8922M! It works perfectly. The spec-an is not that great, it has a max span of 10 MHz. It's missing FM modulator, but still might be useful for transmitter tune up. Maybe some software can make it more useful. This one is kind of a disappointment as a service monitor. Maybe with some BASIC programming it can sweep a duplexer.

    So I started looking at the 8924C which arrived later. No power on! I cleaned it out, removed the power supply and opened it up. Connected the PSU to mains, there should be a green LED, no green LED. Applied hair dryer. Green LED! Two small lyt caps on the daughterboard next to the LED, and it's all fixed up.

    Some dildo had broken off a BNC stinger into one of the input ports. Then pushed another connector into it, pushing the contact a half inch back into the input module. I was a bit afraid to open the input module but it was easy. Unsoldered the connector from the top, pulled the contact out of the bushing and got a razor blade to get the stinger out. Luckily the contact PCB pin makes a hairpin loop so this movement did not fold and break the pin.

    This machine works below the rated 10 MHz too (even with the electronic pad switches, only the levels go down.) So it was kind of easy to use WWV to adjust the ocxo.

  9. #8
    Join Date
    Nov 28, 2012
    Location
    Australia
    Posts
    226
    Thanks
    162
    Thanked 79 Times in 48 Posts

    Default

    You got lucky with the 8924C, good work. For the price you picked them up for you have to be happy.