I just got an HP 8660C signal generator with 86632B modulator and 86601A RF sections. Two quick questions:

1. Any chance of an operator's manual? I did some searching and turned up nothing.

2. It appears to be missing a fan in the back (second photo). Will the unit operate absent the fan, or do I need to hotwire something? I did a quick power-up with metered variac, and the modules appear to power up (needles move), but the main numeric display is dark, although the main power switch works.

Thanks for any clues to this mysterious (and HEAVY!) beast.

Regards,

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html

That was quick. Thanks!

Phil Nelson

You can get the manual (gratis) from HP here;
http://www.keysight.com/en/pd-8660C%3Ae ... =US&lc=eng

The plug-ins each have their own manuals.

That generator absolutely MUST have a working fan that blows a bunch of air.
There's a lot of TTL logic in that thing, and it runs hot.

I have an 8660B system, and I love it.

- Leigh

When I got my 8660 it appeared OK but on power up one of the lines was low (12 v +?). I traced it to a shorted 22uf on one of the boards. Then it happened again; then again. So I ended up replacing every 22uf cap in it.

I have some muffin fans lying around and can rig one up to blow in there for the time being.

Meanwhile, I'm not at all familiar with this unit. Shouldn't the mainframe numeric display light up when I switch from Standby to On? I'm only seeing a couple of lights on that panel (Hz and FM Mode) and it's not responsive to any buttons or controls.

Phil

Hi Phil,

Yes, it normally shows a frequency of 1.000000 MHz when first turned on IIRC.

Sounds like you have a power supply problem, most likely failed electrolytic cap(s).

The power supply has four large can electrolytics, all of which are suspect.

The manual I linked above has full-width foldout scans that are much easier to use than chop-ups.

It also has a very complete troubleshooting tree. I suggest you use it.

=====

The original fan is rated 115 CFM at 115 VAC.

You'll need to measure and match the mechanical dimensions.
That info is not in the manual.

- Leigh

OK, time to haul out the old cap checker and dig in.

Thanks,

Phil Nelson

DO NOT run that with out the fan. The control head is FULL of 54/74 series TTL devices and the heat will ramp up pretty quickly.





This is my spare parts control head.

I believe the 8660 is a "state" machine in that the functions are handled by discrete logic elements, configured into various "states" by the front panel controls, and there is no microprocessor control. I think it was one of HPs last non-microprocessor signal generator.

Wow, that looks like a nightmare to troubleshoot without an extender cable.

The four pin power connector to the digital control unit (DCU) is notoriously intermittent. Put a slight bend in each of the four pins to guarantee a solid connection.

Fans are not expensive and the 8660's MUST have one.

You do not need any special cable to troubleshoot the DCU; all you need is a piece of cardboard (insulator) on the top of the low 20-30 MHz section and, place the upside down DCU there, and reconnect the ribbon cables. You will need two extender boards, be sure that these extenders have independent sides with no thru holes.

The highest failure in the DCU is A1A4U10; I don't recall the failure mode even though I serviced at least 700 8660's as the factory service engineer but haven't touched one in 23 years.

Hi Pete,

Then you'll appreciate the mod I made to my 8660B:

I replace ALL the 74-series ICs in the control module with 74LS.

It really reduced the heat in the module dramatically.

It was a lot of work, but cost very little money.
Since I was given the 8660B for free, I ended up with a really nice generator for pennies.

- Leigh

Phil-
You can use any RF section plug-in since the 11661B Extension Module is already installed. A common failure in the power supply is the relays; you can test them by measuring across the pins on the bottom of the main power supply board. If you find a bad relay you can simply solder a short across the socket pins. Another way to test the relays is to just swap the two around and see if your failure symptom changes. If you smell something burning in the main supply board, remove the filter caps and look for a fried trace; that's what happens when one of the rectifiers shorts.

Leigh-
You must have gone through several soldering iron tips! Luckily it was only the B's that had the 74 IC problem.

Thanks. I haven't done anything yet except skim the manual and stare at this hulk on my workbench. It looks intimidating, offhand. Sort of like, "Oh, I'll just give this aircraft carrier a little tune-up." Yeah, right.

Maybe printing the manual will help motivate me. Not easy to read on a computer screen.

Phil Nelson

I buy original manuals for all my test gear, and HP is especially good documentation. You will find a trouble shooting chart in there somewhere, which will be very helpful. I consider their cost as part of the equipment price.

They take up a bit of room, but...


...the fold out schematics and drawings are invaluable.

PLEASE PM me before you make any adjustments. To align it properly you need a 5 GHz counter or wavemeter, and a 2 GHz spectrum analyzer. If you start tweaking things you could easily mess things up. The counter must have an external time base input (10 MHz).

Don't worry, I'm not close to making any adjustments. Today I installed a new cooling fan; in case anyone needs one, a Dayton 4WT46 fits perfectly and delivers 115 CFM.

It doesn't come one when I power up the mainframe, but nothing else does, either. (Is the fan supposed to run constantly?)

Tomorrow I'll try to figure out how to access those big electrolytic cans in the rear.

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html

I don't have this model, but all the other HP and Tek equipment I have run the fan continuously once the power switch is turned on, and I am sure it its the same here.

I would not mess with those electrolytics yet. Figure out what is causing the unit to be dead first. Chances are that those electrolytics are good as HP used the best money could buy. I learned my lesson not to mess with the caps in HP equipment unless they actually test bad.

I do have a 8660 and agree with what Scott said. The fan is run off the line AC rite behind the on/off switch, so my first guess is the voltage selector switch, the on/off switch, or the fuse.

Are you sure the fan spins freely as installed?

Maybe there's some life in this old nag, after all.



The numeric display lit up after I pulled and reseated this card (A5?) in the back:



I noticed some test points on top for -10V and +20V, and measured the correct voltages there.

The only other thing I had done was to loosen the control head and slide it out a little, in hopes of learning how hard it might be to reach things like the power switch and keypad contacts. Possibly that card in the rear had some funky contacts, not surprising in a high-mileage unit like this.

I haven't found any sort of voltage selector switch. The schematic shows a voltage selector card, which I guess is buried somewhere out of sight.

The keypad is unresponsive and there's still no life from the fan. (It does spin freely, by the way.) I'll pull the fan back out and investigate. I could have sworn I measured 117VAC or thereabouts at the two board terminals where it gets power.

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html