HP8640B -- Fault Found! A very TINY and Hard-to-Fix Fault

 

I've been troubleshooting the internal counter in my HP8640B signal generator. The generator itself works fine.  And the counter works fine for all signals coming in on its "external" port.  But the internal counter stopped working properly above 16 MHz. So I started digging into the manuals and the schematics, re-familiarizing myself with the digital logic behind pre-microcontroller  frequency counters.  NOTE:  If you are working on one of these, be sure to be using the correct version of the manual and schematic. At one point I found a really nice high def copy of an HP8640B manual (for one used in a Patriot Missile System!) only to discover that MY A8A3 board was significantly different from the one in this manual.  BAMA provided a manual that matched my device:  http://bama.edebris.com/manuals/hp/8640b/

Every dark cloud has a silver lining.   Here, the silver for me came from opening up the HP8640B.  What an impressive looking piece of gear.  It looks like something from the Apollo command module, or perhaps from a nuclear weapon.  "The RF source is a 256 to 512 MHz cavity-tuned oscillator that is mechanically tuned..."  There is phase lock circuitry.  The are AM and FM modulators and a really useful array of attenuators.  There is a frequency counter with an external port and an internal frequency counter that measures the original 256-512 MHz signal, then divides down to give a very accurate readout of the output frequency.    This is the kind of device that would generate a cult following.  Count me in! 

Also,  I've sometimes lamented the lack of VHF test gear on my work bench -- the HP8640B could really help me move me into the VHF range.

I started the troubleshooting with some observations and noodling. At what frequency did the internal counter stop working?  What did the readout look like when it stopped working?  What device failure could lead to these symptoms?  I was aided in this by suggestions sent in by readers of my previous blog posts.  Thanks guys.  

I was just getting ready to start some intrusive testing on the logic devices in the internal counter when Dave VE3EAC sent me this: 

-----------------

I think you might be overthinking the failure mode here. I had a similar problem with my unit and it was one of my early Covid-fix-it projects. There is an assembly that controls the bands on the front. It has the famous gears that crack. On the back side are two sets of rotary switches that control a lot of stuff. The switches are of a very unique HP design and offer a lot of advantages over traditional switches EXCEPT they fail in an unusual manner. A PC board has all of the interesting wiring and very tiny double leaf springs short tracks together as needed. The springs tend to break away from the plastic posts on the rotating plate and not make the needed contact. Very carefully examine the insides of the 8640 and your bench top to see if any have fallen out. These are difficult to buy or fabricate. The disk is designed to be rotated 180 and use a new set of posts to locate the springs. Use a small dab of epoxy to set in place. The totally mechanical repair fixed my unit that also would not read above 16 MHz. There are a number of web pages that give great detail of this repair. Also it is worth while to replace the Delrin gears if they are cracked. Replacement brass ones are available on eBay and they will permanently fix the gear problem.

---------------

I had thought about the problem being in the frequency range switch, but I had sort of tested for this by slowly rocking the switch through various positions as I watched the display.  A dirty rotary switch will usually allow the circuit to intermittently work as you rock the switch.  But this didn't happen.  And the HP switch felt quite sturdy, so I focused on the circuitry.  

When I got VE3EAC's message, I carefully flipped the HP8640B over and for the first time opened the bottom of the compartment.  The bottom view is much more impressive than the top view: 

The switches that VE3EAC wrote about  are just below the ribbon cable near the center front. I could see the little springs that he was discussing on the switches.  They appear MUCH more delicate than the rotator on a standard rotary switch.  And I didn't see any of them lying around below the switch.  But when I tried to flip the HP8640B over, something in there moved and caught my eye.  I pulled out some tweezers and pulled this out: 

Wow.  That little spring contact fell off the switch.  That was preventing the HP8640B internal counter's time base from changing as I went above 16 MHz.   It is ironic that such a big and solidly built device such as the HP8640B should be laid low by such a TINY part.  

This gets me back to my original question:  Discretion or valor?  Getting that spring back onto that switch will not be easy.  VE3EAC sent me this K6JCA link describing how to do this.  Yikes, it even requires the purchase of a special tool!  

http://k6jca.blogspot.com/2018/09/repair-log-hp-8640b-rotary-switches-and.html

I'm going to let the HP8640B sit there with the cover off for a while.  It will be taunting me, challenging me to fix it, to make it work the way Hewlett and Packard intended.  It may take a while, but I think I'm going to have to accept this challenge. I've become  real fan of the HP8640B and it would be a shame to leave it wounded like this. 

HP8640B Counter Repair --- Discretion? Or Valor?

 

Inspired by BH1RBG, I cracked open the HP8640B to have a look at the counter circuitry.  Above is the view that greeted me.  That is the main counter board after I pulled it out of its socket.  You can see the seven little red LED display modules.  

It is not as bad as it looks.  In fact, I found the construction and accessibility of the HP8640B to be quite impressive (much better than the Tek 465 with all its flaky plug-in transistors). The manual has good, detailed info on how to get into the various compartments, and even as you work, instructions on which screws to remove or loosen appear on the tops of each RF-tight compartment. Nice. This thing was obviously built with the needs of a future repairman in mind. 

Above is that same board flipped over.  Again, not as bad as it looks. 

Above is board A8A3, the board that I suspect is causing me trouble.  When I go through the troubleshooting routines in the manual, I get to the point where they check decimal point position.  Everything is fine UNTIL I GET TO 16 MHZ.  Then the decimal point is not where it should be and the frequency displayed is very wrong.  (This is in the internal mode -- the counter works fine with an external signal source).  The manual then sends me to Service Sheet 15 which points to possible problems on this A8A3 board, U3, U7 or U6D. 

Troubleshooting this will be tough.  I do not have the extender board that would allow me to test this A8A3 board with the other counter board raised up above it and operating.  BH1RBG noted that getting the extender boards is almost as tough as getting the HP8640B itself. 

While it really bugs me (!) to have  a part of this device not working properly,  I could just leave it as is.  The signal generator is working fine, and I could use the external counter input to check the frequency.  But this is a real kludge. 

What do you folks think?   Fix it or leave well enough alone?  Discretion or valor?  Anyone have an extender board?  Any ideas on where the fault might be?    

The Agony of Troubleshooting -- From China

 

I am troubleshooting my beloved NYC HP8640B Signal Generator (thanks to Steve Silverman and Dave Bamford).  Some of you may wonder why I don't just replace this beast with something small, lighter, cheaper and newer.  Well, I have not found any new sig generators that will do what this beast does:   It goes all the way up to 256 MHz (higher with an extension kit).  It has a great attenuator in it so you can set the output just where you want it.  AM or FM modulation.  Really useful.  So I think I'll fix it. 

My problem is that the internal freq counter stops working above 16 MHz. Using the very extensive  documentation, and without even really opening up the machine, I think I have located the fault.  I think it is in the Counter Time Base Assembly board A8A3.   Now of course, the fun begins.  (Tips, advice, solutions, and words of encouragement would all be appreciated.) 

While planning my assault on A8A3, I came across the web site of BH1RBG.  He too has recently been working on an HP8640B.  His problem was different, but when I read through his site I saw evidence that the agony of troubleshooting is something that is the same all around the world.  Check out his description of the agony: 

I even suspected the LM723 should had something bootstrapping circuit, make sure the Q4 sure start. Because the external reference VR3 is floating too, oh, my godness.

I became hopeless, and ordered several LM732,and waiting delivery for days. This beast frustrate me so deeply, changed the LM723 does not help anything. And i almost desoldering everything in the board!

Oh man, I've been there. Several times while in the throes of a troubleshooting battle I have actually had dreams of removing all the parts from a troublesome PC board.

BH1RBG has a very interesting site with lots of ham radio projects: 

https://sites.google.com/site/linuxdigitallab/rf-ham-radio?authuser=0

Dalibor Farny: Making Nixie Tubes in a Castle in the Czech Republic

“Our customers are interested in technology – some people buy paintings for their wall; our customers buy a technical piece of art. I think they appreciate the fact that someone is keeping old technology alive and they want to support us."

Dalibor Farny is manufacturing Nixie tubes and devices that use them.  He is working out of a castle in the Czech Republic.  Above is a video about his renovation of his workshop.  It was nice that he involved his kids in the project. 

This article tells his story: 
https://hackspace.raspberrypi.org/articles/meet-the-maker-dalibor-farny

Here is his website: https://www.daliborfarny.com/

YouTube Channel: https://www.youtube.com/channel/UC0IY1BQiMehWMvezqWLyk4g

Builds Oscilloscope at age 12! Nick has THE KNACK

Thanks to Grayson Evans for sending us this very encouraging video.  
It is from Tektronix;  I wonder if Alan W2AEW provided some ideas and inspiration? 

More stories like this:  https://www.tek.com/stories  

Twenty-nine S-38s on Craig's List

This is almost like a nightmare.  I guess it could be worse -- they could all be E models. 

https://boston.craigslist.org/gbs/ele/d/arlington-twenty-nine-29-variouss/7188621508.html

Thanks to Jim W3BH for alerting us to this, uh, opportunity. 

Marc Verdiell Has The Knack

Here is Marc Verdiell, the "Curious Marc" who repaired the Soyuz Clock (shown in yesterday's blog post).  

Really cool.  Many SolderSmoke fans will completely understand Marc and his passion for 
electronics. 

Taming Glitches in a Soyuz Space Clock (Plus -- Inside a Logic Chip and How Crystals Work)

Thanks to Bob KD4EBM for sending this.  

There is so much good info in this video:  They crack open a logic chip and look at the internal construction (it is entirely understandable by mortal minds).  They use cool test gear to troubleshoot the clock from a Soyuz spacecraft.  They explain very clearly the series and parallel resonances of quartz crystals, then display these resonances on a very nice spectrum analyzer.

The creator of the video is CuriousMarc.  He has many other interesting projects: 

https://www.youtube.com/c/CuriousMarc/featured

https://www.curiousmarc.com/ 

His bio is here: 

https://www.curiousmarc.com/about

One paragraph from his bio really resonated with me: 

Working on this old stuff forces me to deal with the very fundamentals of electronics (and electro-mechanics). The principles are exactly the same as today, but nothing is hidden in mysterious circuits - you can understand and fix everything. Years of Moore's law has sure given us gobs of transistors, oceans of memory and a glut of gigacycles, but many times, particularly in consumer hardware, these are simply used to cover up poor and inefficient designs - and resource devouring software. What Intel giveth, Microsoft taketh away, as they say. It often irritates me that my PC takes several seconds to react to a simple command, in which time it must have executed billions of unnecessary instructions and consumed a few gigabytes of memory, no one knowing exactly what for anymore. Nothing like this in old high-end hardware: designs are pure and efficient, and the lack of resources is compensated by engineering mastery and immense cleverness, which is a joy to reverse engineer. Not only does it teach us timeless electrical fundamentals and engineering tricks, but it also gives us a much better appreciation of today's tech. How did all the technology we take for granted came to be? It will make you a far better engineer and inventor if you take the time to be a thorough student of the inventions of your illustrious predecessors.

Global Collaboration: The uSDX -- A Multi-Mode QCX

Bill:

There is a new open source, home brew, multi band, multi mode QRP transceiver that grew out of the QRP Labs QCX. Through some serious magic it retains an efficient class E RF amplifier for sideband and digital modes. It crams impressive SDR capabilities into an Arduino. More info at https://groups.io/g/ucx/topics

The basic work appears to have been accomplished by Guido Ten Dolle PE1NNZ. It uses pulse width modulation of the PA supply voltage to transmit  modes other than CW while retaining class E efficiency and uses a direct conversion SDR receiver.

 https://github.com/threeme3/QCX-SSB .

There are several variants by different developers. I built a variant designed by Barbaros Asuroglu WB2CBA  

https://antrak.org.tr/blog/projeler/usdx-an-arduino-based-sdr-all-mode-hf-transceiver-pcb-iteration-v1-02/ that uses through hole components (mostly) and I'm pleased with it's performance. I also designed and 3D printed a case.

This has an interesting development process with contributions by many, including the usual gang of suspects: Hans Summers, Ashhar Farhan, Manuel DL2MAN, Kees K2BCQ, Allison KB1GMX and Miguel Angelo Bartie PY2OHH. I apologize to the many others whose names I didn't list.

The band switch multiband version by DL2MAN is even smaller, but with SMD components which I wasn't ready to tackle yet.

BTW - your podcast encouraged me to go in this direction. I built a BITX 40, a uBITX (sent you a pix of it in an old Heathkit Twoer case), U3S, QCX and now my first step from kits to built from plans.

73

Bob   KD8CGH

https://antrak.org.tr/blog/projeler/usdx-an-arduino-based-sdr-all-mode-hf-transceiver-pcb-iteration-v1-02/

Music Video -- "The Spirit of Radio" by Rush

Thanks to Thomas from The SWLing Post for putting this on his blog.

Andreas Spiess (The Guy with Swiss Accent) -- Workshop Tips

Thanks to Paul Taylor VK3HN for alerting me to the YouTube channel of Andreas Spiess.  There are lots of great projects there, including several videos on building an Oscar 100 ground station.  

In the video above, Andreas talks about his lab/workshop.  He is more focused on digital projects than I am, but I found many of his tips applicable to the analog world. And of course the more digitally oriented readers will find Andreas's observations especially useful. 

I was sold on this video when Andreas reached to his book shelf and showed us Tracey Kidder's "Soul of the New Machine."   That is the book that brought the word "soul" into the SolderSmoke lexicon. 

There are many great videos on Andreas's channel.  I found his Playlists page to be a good way to see the many different catergories of his projects: 

https://www.youtube.com/c/AndreasSpiess/playlists

Thanks Andreas. 

A Song Fom Dexter ZL2DEX

From a post on on "ZL Amateur Radio & Friends,"  in response to a complaint about the lack of activity (with apologies to Paul Simon):  

'50 Ways':

"Get on the band, Stan,
Trigger the mic, Ike,
Tap your key, Lee,
Listen for me.
Paddle the bug, Doug,
Watch the waterfall, Paul,
Just QSO, Joe,
when conditions agree."

73

Dex, ZL2DEX