I read as many old threads as I could before this. Waded through tons of web pages and some of that is summarized here.

I've installed a 20dB 100 watt attenuator in my 8924C and need to load the matching cal files.

Now, down the rabbit hole... looking for a solution for loading 39 kilobits of calibration data into the 8924C. The files are available here on the forum. For loading in cal data, the 8924C can look at the following locations: external HPIB disk drive, Internal ROM, Internal RAM, external front panel plug in SRAM (or ROM) card. External disk drives are scarce. Unsure how to interface such an old disk drive to a modern PC or how to load the cal files into a HPIB disk drive (or to load into a PCMCIA card). There are some web projects out there to emulate the HPIB disk drives. That's a bigger and deeper rabbit hole.

Memory cards background info - not the same as the slim, 40 pin card as used in the portable instruments such as 85xx series spectrum analyzers or the 892x portable service monitors.. The correct cards for the 8924C are sometimes called "Epson memory cards". Based on a 68 pin 16 bit or 32 bit CardBus interface for PCMCIA cards. Loading data into them can be difficult from current generation computers. Amtronix no longer sells the cards that were part of their 60w and 100w upgrade kits with the attenuator assembly and new front panel sticker.?

I think a CardBus type 1 is 16 bit only and is the slimmer 3mm card. It might be what the HP 16 kB and 64 kB cards use. Cardbus type? 2 is the 5mm card and can use 16 or 32 bit addressing.

A few HP SRAM cards can be found on Ebay for big prices. I don't know if a combination of a "small" 32 Gb Compact Flash card and a CF card to PCMCIA bus adapter might work.

Or maybe somebody out there has cards to sell?

Thanks!

Jim
?

The biggest problem with the 8935 power supply is the open 47uf/50V capacitor on the daughter board. You need to remove the rear power supply, remove the first circuit board and the main mother board is under it. You will see several daughter boards. Locate the "daughter board" between the R24, R25 resistors. The 47uf is on that board. Also known to fail is a 4.7uf cap on the daughter board that has the (2) potentometers.

Good luck, Rick_,_._,_

HP 8920 power supply help:

Here are the main problems with the HP 8920 power supplies:

If your supply fails but the fuse does not blow: the first thing to check is the R3 resistor it's a 470 K ,1/4 watt. I recommend replacing it with a 1/2 or 1 watt.?

Next is the rear AC/DC power switch. For some reason, that switch goes open a lot. Use an ohm meter to check the switch. If it's open, replacing it is difficult because it's difficult to find. You can always remove the switch and jumper the connections to make it in the "AC" position all the time. Be sure to remove the switch first because if it's accidentally pressed it could damage the supply by having it in AC and DC mode.

If your supply smoked but is still working, that's generally an electrolytic cap that shorted. I recommend replacing all 9 filter caps (5 - 1000uf/35V, 4 - 470uf/50V) 2 - 47uf/50V and C8 - 33uf / 50V. I also replace the 630V film cap (C7). These caps are getting old and known to leak on the board.

If the main fuse blows, it's generally the Q1 power FET that's shorted. Sometimes when Q1 fails, it takes out some resistors. Also check the diodes.

开云体育

Make sure to check resistor R3 for OPEN...it will keep Q1 from working......it may look just fine but be open..

开云体育

No schematic attached..could you please attach one...thanks..

Hi,

I have the schematic of the 8920 model. It is not in the normal download of the CLIP. It was manufactured by an outside company for HP (Artesyn), along with the Spec An Module.

I have attached two PDF Files.?

Enjoy your HP. Thanks to those who have been there before you.

Regards

Bill Macham

Perth Australia

?

?

I've never found schematics for 8935 series. Closest you can get is the HP 8920 CLIP (Component Level Information Packet). Some of the modules are same / similar but power supply is way different (but also suffers from dry caps)..

73 de ZL1CVD

Can't locate any schematics for the 8935, nor a parts list for the power supply. Anyone ever compile a parts list after recapping their 8935 power supply?

TIA,
Bryan K6RBJ

Rick, thanks for the reply. Yes, I am aware of that message. I removed them (due to age) when I took the back cover off to check the LED status on the motherboard. I just haven't gotten around to replacing them yet, I doubt those being missing would throw out the 40 volt failure we also see on that screen, but I suppose anything is possible.

The 40V supply is crucial as it is used in various PLLs for VCO PIN diode tuning voltage. Have a look in the Agilent 8935-E6380A Assembly Level Repair Guide page 78. I bet several modules have the OOL LED illuminated. The fix is to pull the power supply & replace all the electrolytic capacitors. If you need a quick fix, contact and see if they can supply a reworked power supply. Note you could just track down and fix the faulty caps in the 40V section of the supply but when I did that in the past, I had to go back in a few months later to replace other dry / crook caps.? Its better to do the whole lot in one go. Oh and I can't recall if this model had a separate regulator board or not. If it does, change the caps in there too. Now once you change the caps / power supply, make sure you do a smoke test when you power back up. I have had several tantalum caps blow out when the power supply was repaired. They are generally pretty easy to locate by watching which module is smoking :)

The screen says your AA memory batteries are low. I would test them and replace if low. If they measure good, perhaps there is a problem with the A/D conversion that measures voltages.

开云体育

Hi Bryan,

?? In the unit on the mother board are 4 green LED's that should all be on. They are the +,- 5Volts, 12 and 40 volts. I would check them. Rick

Hi,

I've built a GBS8200 converter with the well known LM1881 circuit.

There is a clever design available for a 3D printed (FDM) enclosure: (provides BNC slots, originally seems to be intended for RGBS via BNC or SCART to BNC inputs).

Unfortunately, I have some woes with the converter. It reports 'no signal' in RGBS (or the other H/Vsync RGB mode), but I can see at the BNC composite video input the signal is present. I followed Ralph's wiring as detailed in /g/HP8924-Family-Spectrum-Analyzers/topic/82180602

Pending some proper diagnostics with the oscilloscope, my protoboard with the LM1881 has decoupling capacitors (100nF) at VCC and P2 and P6 as described by Ralph, and doesn't seem like there is anything obviously wrong with it.

I quote Ralph's wiring notes verbatim:

On the LM1881 connect pin 1 to the # 4 (gray wire) pin of the 8 pin RGBS input connector.
Pin 2 to a .1 UF capacitor and then to ?the Green input of the 8 pin RGBS connector?
A wire from that junction goes to the input from the 8924C.
Pin 4 to ground.
Pin 6 has a .1 UF and 680 K resistor in parallel to ground.
Pin 8 goes to 5 volts.
Pins 3,5,7 are not used.

I crimped connectors to make a harness so I can unplug the LM1881 board if desired. Green for RGBS is crimped together with the pin 2 trace of the LM1881 (composite video input, per )

LM1881 might also be replaceable by the more modern LMH1980 which can operate at 3.3V. This is quite convenient since the GBS8200 has regulated 3.3V at the 0-ohm jumper/resistor labeled R1, and it's sufficiently filtered through decoupling caps and at least one smoothing electrolytic cap.

Does anyone have close up pictures of their LM1881 board? Is there anything I am missing? Could my 8200 be defective?

I performed the modification with the 50k resistors at the RGB traces with the ASIC/FPGA under microscope and I know there are no bridges or solder wicks anywhere.

?

Cheers

based on what screens you have included the first thing I would look at is your generate output level.? Suspect the counter failures are due to a low level and the MISC DIAG is due to the same.

Look at the Duplex out port ...set the level out to something like 0dbm? at the DUPLEX port and make sure it matches with an external SA or power meter.

If it doesn't match? switch to the RF I/O port? and set the level to -20dbm and see if that matches.? If the RF I/O is good than one or both of the? DUPLEX? and ANTENNA ports may have issues ( blown protection diodes ) but the DUPLEX port would be first item to check.

If both are low than it may be the generate module.

chances are? the DUPLEX level is? off but it may be that your antenna in port is blown or you have a weak? MMIC in the RX? RF path .... the full screens would help.

Normally on these units I pull the RF I/O and remove the cover to check for blown input protection devices on the DUPLEX and ANTENNA ports.

Given the counter failure it leans more toward a week MMIC? but you may have multiple issues like a blown port or other issues

You may want to check your RF cable from the backplane to the RF I?O module ...re-seat the backplane end ( 5/16 wrench ) and tighten the connector to the module ...this is close to the power supply and those are often fun to remove and the cable may have gotten loose.