Counting on the HP 5328B

[jrharley]

Greetings fellow test equipment junkies!

Recently picked up an HP5328B Universal Counter. Serial Number: 2450A00594. Options 010 and 031are present. Pretty clean unit overall. As is often the case when buying used gear, this one arrived a little worse for the journey. The seller packed it reasonably well, but not well enough. Even though the box showed no signs of it, it obviously took a vicious tumble at some point.

One of the cast aluminum rack handles was broken clean off and the unit no longer had a display. Checking the rails on power up showed a missing
-5.2V rail. All boards were subsequently removed and examined. All card edge connectors were scrubbed with a fiberglass brush, cleaned with IPO and treated with a light brushing of D-100. They were re-seated into card sockets cleaned with CRC and treated with D-100. The missing negative 5V rail cleared after lifting a few likely components on the motherboard and re-installing them.

Upon re-assembly things started coming back to life. Display reads a solid 10.00 MHZ in the check position and the unit is responding to switch manipulations. Still lots of work left to do to run it through the maintenance checks etc., as well as some cosmetic items. The most troubling of those are the Function and Freq. Resolution knobs. Some genius decided to seat the grub screws into what appears to be some JB Weld, and so those guys are not coming out without a drill bit. Pondering that situation for a bit.

So, I’m calling out for any options folks can suggest for knob sources, as well as feet for this unit. Looks like standard HP feet will work on it. As always, any thoughts, advice, etc. are appreciated on this guy. A photo is attached for reference. I’ll report back on functionality after I get into it for awhile. Thanks in advance for any help!- JRH

[Stray Electron]

   It's too bad that the Chinese or someone like them haven't started reproducing the HP feet.  Used TE always seems to be missing at least one of them.

   The knobs on much of the HP gear are surprisingly unique to each model!  I've collected a pile of them and even though some of them look very similar, when you compare them closely you'll find differences. 

    My strong advice regarding the knobs on yours is to buy a set of left hand drill bits and use on of them to drill out the JB-welded set screws. LH bits will frequency unscrew broken or stripped bolts but I don't know if it will get out the JB-welded ones.

    The best way to handle set screws that have tendency to loosen up is to simply install a 2nd setscrew on top of the first one.

    One other thing, some of the HP knobs use 1/16" set screws (0.0625") and some use 0.050" set screws so be sure that you're using the correct size wrench.  With either of those sizes it's very easy to strip the both the Allen wrenches and the screws!  Buy a good set of small Allen wrenches and put them somewhere safe and take good care of them.  Check the corners before you use them, especially when dealing with tight screws.  If the corners of the wrench start to get rounded, the wrenches can slip inside the set screw and round out the hole in it and make them very difficult to remove.  I grind the tips of my wrenches down when they start to get rounded so that my wrenches always have a good sharp corner at the end.

   FYI, you called them grub screws but we refer to them as set screws. 

[tautech]

3dp models of HP feet are here:
https://www.eevblog.com/forum/testgear/replacement-knobs-feet-and-fittings-for-test-equipment/msg2255118/#msg2255118

[jrharley]

Thanks for weighing in Stray! Good to know you’re on it.

All good points. Yeah, I’ve laid into one with a left hand bit, but there’s no backing these puppies out. They are seriously bonded. Since these cool “Thumb Paddle” knobs are composed of 99% Unobtanium, it seems to me that preserving them, even if the, ahem……SET screw core, is sacrificed, should be the prime directive. Bottom line in this situation is that I need to sleep on this one for a bit. These set screws are the .05 versions. Tiny little Grubs! 😊 Looks like I may need a source for those too. You’re right about Allen wrenches, thanks for the tips! I’ve got more Allen wrenches than Allen himself, but haven’t thought much about the details. I keep one brand new set in the lab for precision use only.

Thanks for the links tautech! Looks like some browsing is in my future.

Really appreciate you guys chiming in!  -JRH

[factory]

It shouldn't be hard to find grub screws in UNC sizes where you are, you have places such as McMaster-Carr;
https://www.mcmaster.com/products/screws/fastener-head-type~headless/set-screws~/thread-type~unc/

In the UK I've used Accu, choice isn't as good of course;
https://www.accu.co.uk/shop/precision-components/precision-screws/grub-screws/socket-head-set-screws/imperial-cup-point-setgrub-screws?page=1

The common sizes of grub screw & allen keys used for HP TE, the #8-32 is usually found on much older HP TE;
#8-32, 5/64"
#6-32, 1/16"
#4-40, 0.05" (50 thou)
#2-56, 0.035" (35 thou)

For removing the grub screws I usually tighten them first to break the corrosion, it works for me most of the time, if it's too far gone drilling is probably the only option. Tek stuff seems worse with steel grub screws into aluminium.

David

[jrharley]

Thanks factory! This is helpful and will save me time, appreciate it. 

I’ve decided that the screws are no longer the issue in this situation. We have metal screws bonded into their original threaded holes with a very capable epoxy- Full stop.
Gettin’ all medieval on these guys with a hand held drill is not going to yield satisfactory results. (Not that anyone has suggested that, just sayin’) The attached photo illustrates some of the dimensional challenges that exist regardless of the technique.

What is needed it seems to me is a vertical, well restrained drilling condition, that removes the screw as precisely as possible, accepting that the plastic threads are going to be compromised in the process.

Any way I look at this problem, the potential to “Kama-gatza” on this thing is very high. I’m not in a hurry to solve it. The knobs are functioning and there’s stuff to do with the operation of the machine. For the time being, I’m going to focus on that aspect. As always, I really appreciate the feedback folks.- JRH

[jrharley]

HP 5328B update here:
So, I’ve been going through the maintenance checks on the unit. When the function switch is set in the “check” position, the readout is 10 MHZ, across all the resolution settings. Pretty solid there. Crystal oven warms up nicely.

It’s clear we have an issue though.

Check #1 – “Time Base Stability & Output”
Epic fail. All I get is 00.000 on the display. It doesn’t read the 1MHZ signal being applied from the sig gen. It would seem the timebase is not so stable. 
Just for grins, I hooked up a decent, battery powered 10 MHZ signal to the external oscillator BNC on the back and switched it over. No change. No input signals register.

Check #2 – “Self Check”
All good with this one. The unit responded as expected and described in the manual.

Check #3 – “Frequency response & Sensitivity”
Nada’. No readout except 00.000 during this procedure either.

Nope, she’s not reading anything from the outside world it would seem.
The documentation I have been able to get is not that great for this ‘gal, but I’ve got some schematic study ahead of me for sure. i'll post up with any developments as they occur along the way.

Since this Bodger is, “skiing a slope he has no business being on”
I always appreciate any tips, thoughts or suggestions about the project. -JRH

[Tony_G]

Just on your point of Check #1 - I remember hearing that the Ext OSC input is sensitive to the input level - You might want to check the manual to see what the spec is to make sure that your 10MHz source is generating a signal at the required level.

I don't have one of these units so all I can really do is to point at HPAK Groups and watch how it turns out for you.

ToyG

[jrharley]

Thanks Tony G! Appreciate you checkin’ in. 

10-4 on the input levels. I set the input signal levels to the values specified in the manual, and was generally able to set up the tests as outlined. The 3rd check involved frequencies to start with that were too low for my 8656A to deal with, so I improvised with a BK Precision function generator, monitored thru a digital scope, and was able to dial things in as described in the procedures. All was good on the input level for the external feed as well. (I will double check that though.......)
So far, I haven’t been able to find documentation specific to the B version of the 5328 I have. It could be that the B version was just documented with addendums and I haven’t come across those yet. Even if a download indicates it’s for the B, the manual always says it’s for the 5328A when you view the PDF. Much study of the manual is ahead for me I think. I’ve started poking around the board where the inputs land first, and quickly realize that troubleshooting them under power is going to be interesting. Now I grasp the significance of the so-called “extender boards”. I think I’ll just order up a set from the accessories parts list in the manual and give it a go. 

Meanwhile, onward testing voltages and learning what I can learn about this jewel. Thanks again for commenting! -JRH

[Tony_G]

I will keep watching to see what happens.

Extender cards can be made if you don't have them - I did a 44-pin one using KiCAD & PCBWay and bought the connectors from Newark - There is a thread on this over on the HPAK Group.

You may end up needing to spring for the manual from one of those high price eBay sites like here eBay auction: #202259862098

Good luck.

TonyG

[jrharley]

Appreciate the watching TG! Calamity is certain to crop up! 

 I think in the short term, I’m going to work on a cracked, but thankfully not glued knob. It should be a good sidetrack, while I poke and prod the 5328. I find it’s sometimes good for the pondering, to jump to mechanical issues for a different perspective on things.

Speaking of pondering, It’s hard to stomach paying for manuals, but I feel like I need to seriously consider it for this unit. There’s nothing like detailed information. As I talk myself into it, I recall having a local print shop, print and bind an online manual for a scope awhile back and it was at least the cost of the Ebay listings. Getting an original manual does have value.

Thanks for the help! - JRH

[jrharley]

HP 5328B update here – So I’ve been poking around with the incomplete info I have, seeing what I can see. The 5328B version, seems to have incorporated as standard, what was on the A version, an option. The A14 Multiplier board is installed on this guy, but not listed as an option on the back of the unit. The schematic for this board is not in the manual I have, but I’ve been prodding about anyway. There are 2 trimmer caps, 10 to 60pF range at the top of the board. One of those guys seems to be a dead short. The plates on the cap don’t look to be aging very well. I may take it out for a closer look while I wait on better schematics. Both the +5V and the +15V rail are present at the card socket as indicated from the markings on the board.

Meanwhile, I’ve been working on stabilizing a cracked level knob on the B channel. Dialing in on it has taught me a good bit more about these knobs. A couple pics of the journey attached. - JRH

[Tony_G]

Nice work to stop further degradation - If I ever end up at the point of needing to replace one of mine (on whatever piece of gear it is, don't have a 5328B), I'm planning on creating a 3D model and seeing if I can print a reasonable facsimile. Will stick that up on the HPAK group.

TonyG

[jrharley]

Appreciate it Tony_G!

5328A update:

We are counting now!....well, …..sorta’.
Channel A & B aren’t doing s@%!, but channel C is looking good for this stage of the journey. It’s counting across the range up to about 800 MHz with a 250mV signal, which is where I stopped. No worries here. Is it in calibration?.......not a chance, but it’s accurate and stable to about 4 digits. I’ll take it for now. Things seem to be converging on channel A as the place to pay attention to, but this is saying that the oscillator is working fine, along with the 1.3 GHz board. I now have decent technical information specific to this unit. That means I’ll be crawling into a hole to study this stuff. Replaced 3 RIFA caps in the meantime. Yeah, they were all cracked, still measured good. Onward! - JRH

[jrharley]

Happy Weekend fellow TE addicts!
The new manual information is a great help. There's a ton more information in it than anything else I've seen. All of it is very specific to this unit and in general, they did a nice job with the reproduction.

I’m zeroing in on the “synchronizer assembly”, board A10 to begin with. Working on breaking the thing back open to remove the A10 board. It’s not too bad a job, but you can’t just yank the board out. The faceplate’s gotta’ come off to remove it.

A couple random progress photos attached. - JRH

Having trouble with the pics, not sure why

[jrharley]

Hmmmnnn..... trying a couple more. SHIELDED-C.pdf (242.21 kB - downloaded 13 times.)

Even though files are small, only allowing 1 at a time.

[Tony_G]

Good to see that you are making progress - As an aside, to post pictures here I created an Imgur.com account and then just use their BBCode links to include the image and thus there are no attachment limitations:



Look forward to seeing the next update...

TonyG

[jrharley]

Thanks for the tip TG!
Imgur is a pretty disorienting platform to an old guy like me, you know, with all the photo memes and stuff. I'll consider if it makes sense to upload that way for me. Still, it shouldn't be tricky, as a matter of practicality, to upload a couple pics to this thread. I should go back and read some forum guidelines to make sure I'm not doing something stupid.

Meanwhile, I'm deep into studying this beast and learning what I can. Attached photo shows a necessary step.

As always, I really appreciate you chiming in! 

[jrharley]

HP5328B update 'yall, Happy 4th! –

Well, I’ve moved on to the A12 Amplifier board. My limbic dominated brain, reasoned that since that’s where the input jacks connected into the circuit, I should dial down on that board.

Interestingly, I began to notice that a large number of the solder joints, mostly on the topside, looked kinda’ dodgy. The microscope confirmed that. Don’t know if any of them are contributing to not displaying the input frequency yet, but I figure they should be touched up. Unfortunately, during this close observation, I saw a stomach churning flaw in relay K5. The sucker is pretty badly cracked on one end. I don’t know too much about these reed relays, but I’m pretty sure a big split in one end is not a good thing. The coil is apparently 5V, so maybe I can devise a way to test if it’s working using a DC power supply and ohm’ing it out. I sure hope I can determine that it's working, because initial research says it’s not going to be easy to replace. 

Photo of A12 attached. - JRH

[Tony_G]

Quick poke around seems to show that this is a replacement part - Not saying it is but just what the InterWebs seems to point at - Might be worth looking at the datasheet etc...

https://www.littelfuse.com/products/magnetic-sensors-and-reed-switches/reed-relays/he700/he721c12.aspx

There is a 5V version on Digikey here:

https://www.digikey.com/en/products/detail/littelfuse-inc/HE721C0500/133219

Again, not sure it is actually a replacement but might work or lead you to the right part - Good luck.

TonyG

[jrharley]

Appreciate that Tony_G!

The reed relay specs are: 500mA, 5V coil, 1B form, 100VDC, GORDOS part# 0490-1183.
The tough part about these so far, is that they're either not stocked or crazy expensive, or both.
There's an interesting ebay store called F-4 Technologies that carries lots of obsolete technical parts, HP in particular, and they actually have a couple of the exact replacement relays. With shipping, one would be about $25. It looks like GORDOS has been absorbed into another company, but I think I found an old data sheet from the original product line. I'll study that tonight. I also may need to browse some "second tier" supply houses. Something about these specs quickly narrows your choices using the inventory filters. I'd like to find a source where I could get a handful of them for a few bucks each, but we'll see how we go with it. Holler at me if this is stupid, but it seems like I should be able to put a DMM across it in continuity mode and tickle it with 5V to see if it connects. I suppose I should limit the current to 250mA or so.

Anyway, that's where things are. I'm about half way through touching up the solder joints on the board. -JRH

PS - Just noticed that the installed relay part numbers don't really match those listed in the parts list. Maybe they've been replaced, a couple of them don't seem to be seated perfectly.

[Tony_G]

I would lean towards trying to find a current part that matches or is at least reasonably close before I pay those exorbitant prices for exact replacements - Often those prices are calibrated to generate the seller the maximum income from companies who have done the cost analysis of replacing the part with a different one and then redoing "certification" compared to just getting exact old parts.

Look forward to seeing where you end up.

TonyG

[coromonadalix]

i would check the ic socket on your A12 board,   i had some in the past who became weak on contact strenght,  you need to fell  some tightness ...

[jrharley]

I hear you TG!
I set a budget for restoring these old kits, kinda' a limit in my mind before it goes into BER status. I ain't there yet! 
Still, it adds up quickly and it doesn't seem to make sense to spend much on a reed relay. So far though, there's encouraging developments on that front.....
Diving a little deeper into the info on the data sheet, they show schematics of the different relays. It would seem that this one is "normally closed", and has a diode in it to arrest back EMF from the coil. Made sense, so I tested it in circuit, no power applied of course. Diode is present and there's continuity across the relay, indicating to me that those reeds are touching. Power supply is warming up to see if I can get it to open.

Thanks for that feedback coromonadalix! Appreciate it.
Good advice for sure. Hey, if it's not a capacitor, then it's a connection! Amirite? I've spent a fair amount of attention on the card edge socket connections, but not entirely sure, that is what you're referring to. When you say "IC socket", is that the one you mean?

Appreciate the comments guys, it really helps in the discovery. - JRH

[jrharley]

OK- we have a small bit of progress here. 

After tacking a couple small wires onto the K5 relay pins on the back of the board, a DMM set to continuity was connected across them. Since the relay is normally closed, continuity was indicated, annoying the rest of the household. Touching a current limited 4VDC across the same pins, activated the reeds and interrupted the continuity signal. The relay appears to be functioning!  The data sheet schematics were very helpful for the setup.

I do think I'll brush a little UV resin onto the cracked end though, just to seal things up.

With that distraction out of the way, I'll be able to keep studying the schematic and find out why this thing doesn't recognize an input.

Later. - JRH