HP 5065A Rubidium Standard Repair Log

[maat]

I recently got a rare HP 5065A Rubidium frequency (secondary) standard off Ebay (https://www.eevblog.com/forum/buysellwanted/fs-(uk)-hp-5065a-rb-frequency-standard-other-calibrators/). This is one of the old time goodies.

It was a "For Parts" sale and the seller didn't give much info, but it was a rare opportunity to get one of these over in Europe and shipping from the States is a killer due to their weight. So it took my chances, expecting the worst and hoping for the best. The 5065A arrived this week and a first glance revealed that (one of) the previous owners was "Devonport Management Limited". Port? Do I hear sea port? Salt water? Yes, you guessed it, mild corrosion everywhere, but lets do an inventory first.


This is a rather recent 1989 model with all options (that is option 001 and 002). Option 001 is a clock module, which gives us a, well...uhm, clock, a 1 pps output and a sync input. This is a must have for anyone who can't stand it, when the integrated clock of your favourite data logger goes out of sync. Love it! I'll never miss a second for the rest of my life.

Option 002 is a NiCd backup battery which gives you 10 minutes of battery life for a whooping 16 h of charging. We will get to that later.

Being a more recent model it also features the better 10811-60109 OCXO. Very nice indeed.

I don't have a power cord yet, but that is something Mouser can fix. Meanwhile I opened it up and took a few pictures. Unfortunately half of them are still on the mobile phone of my colleagues, so I will post them later. The first thing I was greeted with, was a bag with “loose nuts and screws” written on it – bad sign. The rubidium cell was disconnected and the wires hanging loose. The battery was corroded and had spilled its guts – yuk. But otherwise at least everything seemed to be there. Nothing missing at first glance. The first picture shows the the inner assembly, from which I have removed the battery the A6 board and the rubidium cell. You can see some of the rust I was talking about on the oscillator housing in the right bottom corner.

I checked the boards and obviously someone already had a shot a repairing them. Man, I just love it when people leave the flux on the boards...Some of the boards have new caps on them, but no cleanup was done in the areas which had signs of capacitor leakage. I will post the pictures of the boards as soon as I get my hands on them. All the boards will require new caps and some thorough cleaning to get rid of the copper oxide.

What do you guys think is best? Citric acid or vinegar? Then some sodium bicarbonate (baking soda) to neutralize, distilled water and finally IPA. That would be my strategy and everything that is left will meet my friend, the fibreglass pen.

[Theboel]

Good Luck for repair Your 5065,
if You continue to report the progress I believe some Gentlemen will very helpful.
btw one of the very kind and talented gentlemen already looked at Your thread 

no no no its not me

[edpalmer42]

Mine also had the battery option and it looked about as bad as yours.  I decided not to replace the batteries, but I did keep the plastic battery holder - just in case I change my mind later.

For the power cord, don't bother with Mouser yet.  That military connector is pricey.  Just go into your junk box and find three female sockets that are close to the right size.  Solder them onto a cord, pinch them down so they're a snug fit, and heatshrink them for safety.  If you really want to, use a cable clamp to hold the cable in place so there's no strain on the connectors.  Once you've got the unit working, you can decide if you want to get the proper connector.

Regarding dead capacitors, take a look at this picture:



This isn't my unit, but mine was about this bad.  I almost didn't realize that there's supposed to be a trace under the third capacitor that passes under the two resistors and carries on to the right.  On my unit, that trace was completely gone.  Watch out for that.  I just used vinegar and then water to clean up the boards.  By the way, the picture is from this page:  http://www.niceties.com/atomic.html .

Ed

[maat]

In the meantime I had a look at the lamp assembly. Oh deary me, I see mishandled PoziDriv screws. Just a reminder for everyone repairing HP gear. The usually use PoziDriv for their gear. So check your screwdrivers before digging in! The PoziDriv screws have additional radial indentations at 45° from the main cross. Do not use a Phillips screwdriver on them, ever! (Unless you are disarming a bomb and running out of time, but even then...show some respect for your gear). I can highly recommend Wera screwdrivers with the laser tip (not affiliated in any way). They have incredible amounts of grip even with damaged screw heads.

Back to topic. This is not a good sign, because Mr. NoCleanFlux was probably meddling inside. So I did a sneak peak as well. The lamp seems fine (picture 3), but what is that on the bottom side of the pcb? Can you spot it? Black marks on the mounting spacers. Uh Oh. Reminds me of IT crowd.



And of course more havok from Mr. NoCleanFlux. So yep, he was here as well. Probably trying to clean up the burn marks. See picture 4 to see what it looks like underneath. Again look closely. Hint: the black capacitor.              Crack in the middle. Next to it char marks in black on another component. Ok, I will have to rip that apart as soon as I identified the damaged parts. The red "blob" in the background is another capacitor. Broken as well...So we had quite a hot party going on here. But the good news is, that the lamp looks fine. Phew. Lets see what the vapour cell looks like. I am in luck. No broken glass. Looking good.

[maat]

Yep my boards look mostly like that, with A6 in a little worse shape, which is why I removed it for now. I will check all caps and probably replace them just for good measure.

I haven't decided on the NiCd batteries yet. But I guess I am ever going to retrofit batteries, it will be a lithium ion pack. The charger board A6 is slightly charred anyway. From the schematics it looks like a drop in replacement board with a lithium ion charger should be doable.

[maat]

I finally got the pictures, my colleague took. So here they are:

First the device as received:


Then a leaky battery:

[maat]

The well cooked A2 battery charger board:

Yeah, sure thing. Put 85 °C, 2kh, replacement caps on a board that shows burn marks. 


Backside:

[maat]

The A15 power supply board:

This is the newer version 6100, which was a replacement for the original 6023. Again lots of flux residue. And I will have to replace those blue 85°C, 2kh caps again.


Backside:

[maat]

The A14 logic board:

Pretty decent shape. Some transistor caps are little dull, but overall very nice.


Backside:

[maat]

The A9 Integrator board:

Again this is an upgrade to the older 5060-6015 board. Again flux residue on the OPA111 (https://www.ti.com/lit/ds/symlink/opa111.pdf). Nice opamp, btw. You can still get them. I keep my fingers crossed, that this one is ok...$50 a pop for a new one...

I love that handle they put on the board. This is a ingenious, because the boards are rather hard to pull out.



Backside:

[maat]

The A4 100 kHz frequency divider:

This is one of the boards in worse shape. We are getting closer to the leaky battery. The cap needs to be replaced, as it is leaking as well. Does anybody know the model of the the transformer T2?  It has the HP part number 05061-8008. I don't like the rust on that one. I will desolder it and have a look at the bottom side.



Backside:

[maat]

The A8 phase detector:

This board will require some vinegar treatment as well. Unfortunately one of the caps used in the oscillator part, C3, is a little corroded as well. Its a 25.2 nF, 1%, PS, 50 V capacitor. I will see how bad it is, because this one might be hard to source.



Backside:



Rusty PS capacitor:

[maat]

Finally the A6 1 MHz frequency divider board:

I don't like what I see. I hope the transformers are allright and we will see how the trimmable cap survives the washing. Thats it for now. I did not yet pull out all boards. Got to get back to the kids now



Backside:

[Dr. Frank]

Doesn't look that bad.
Good luck, furthermore.

Frank

[5065AGuru]

Maat,

Thanks for the PIX! The only thing I see preventing an initial power up is the missing Db9 chassis connector for the optical unit. You need to very sure you wire it correctly! As Ed mentioned Make up a power cord with 3 female pins that fit onto the power pins. Heat shrink them. That way you can troubleshoot without having to wait for a connector.
If you are going to work on the lamp assy. replace the 1.33K film resistor as they are a failure point. I use 1.3K 2% 1 or 2 Watt metal oxide resistors. Once happy with the repairs you can test without installing into the optical unit by connecting +20V on the bench and after a minute or so the lamp should light. About 150mA lamp out and drops to about 125mA when the lamp comes on.
Your shot looking into the optical unit shows the Rubidium filter cell. The other end houses the Rubidium reference cell but it's not visible as it's inside the microwave cavity. DONT remove as retuning the cavity is a chore!
Good Luck!

Corby

[5065AGuru]

Maat,

In order you presented.
 
A2 not really cooked, that is a rather clean example. ( As I normally remove the battery option that board goes also.)
A15 has some damage from the 100uF caps but otherwise OK
A14 OK
A9 looks fine the op amp should be OK (during the 5065A alignment if not using the manual/changes that include this board set/check the loop gain on the CW 50 units only!)
A4 I see the pesky 100uF! The transformer should be OK
A8 I'd pull the PS cap, clean its leads, re-tin and put it back in. I can provide a replacement if needs be.
A6 100uF cap again, I've never seen a TX fail, DONT adjust the cap unless absolutely required and if so let me give advice first.
When you get to the A11 replace the two 22uf electros.

Cheers,

Corby

[5065AGuru]

Maat,

While you have the lamp assy removed check and see what condition the reflector is in.
They can get corroded and lose reflectivity. You can look through the Holes in the front disc and also from an angle through the rear hole with the reflector removed.
I have found that "Mothers" Aluminum polish and some elbow grease will polish them up nicely!
After removing the reflector from the assy. and removing the 3 screws holding the perforated disc you will most likely find the disc stuck tight. I put the reflector assy. in a vise at an angle and apply pressure till it pops free. Then push the clear glass disc out of its recess to get at the reflector surface.

Cheers,

Corby

[maat]

Thanks for the hint Corby, I hope that tomorrow I can spare I bit of time and work on the 5065A in the evening

[maat]

So here it is: Part II

Today I took apart at the the (upper) lamp assembly to get a look at the rf generator and that cracked mylar cap (red circle on the left side). I carefully desoldered the wire connecting the the power supply and the board (red circle on the right) and removed the screw pressing the heat sink of Q1 onto the aluminum base plate.



Since there is no picture of the RVFR board in the manual I decided to take one. I am not sure if C3 is small green cap. I forgot to check with a multimeter. Does anbody know, what the two white tubes are (e.g. between L2 and R2)?

[DC1MC]

The little tubes  are small ceramic capacitors in 10s up to 2-300pf range.

 Cheers,
 DC1MC

[maat]

Capacitor massacre

Upon inspection of C2 it turned out that, those mica caps are definitely some sturdy bunch. I desoldered it to have a look at the crack and it turned out, that part of the back side was missing as well. Poor thing, but still hanging in there and within spec






I also pulled C1, because it had some hairline fractures and I do not want it to short out and burn down my lamp assembly. The weird thing was, that C2 was 130 pF instead of the 110p given in the datasheet. Did they tune each circuit by hand? Anyway, since the lamp worked before I pulled the caps. I will replace them with the same value.

[maat]

Cleaning up the battery

Next I started cleaning up the mess the battery caused in its final hours. I removed all the boards to prep them for cleaning and also removed the aluminum housing that contained the battery and the one that contained the boards, Fortunately those two assemblies are the ones most affected and the Multiplier A3 housing will probably get away with some exterior isoprop cleaning.

Starting to look empty:


This is what those parts looked like. All tarnished by the acid.


I gave them a good scrub in soapy water, which removed some of the larger particles. But since the tarnished aluminum was there to stay. I got out the the sandpaper, grit size 600 and I removed just the upper layer of oxide. Using water during sanding helps. Do make sure that you are sanding only in the direction of the original pattern. Except for some places, where the acid has really eaten into the material the battery compartment now looks like new.



I will use the angle grinder and a stainless steel wire wheel to remove the oxide layer in the rest of the case once I have decided what to do with the big electrolytics used to filter the mains. They got spilled with acid as well and show some decent corrosion on their base plate. I will probably remove one of those next to have a look underneath.

[FriedLogic]

I normally used a wire brush with all brass bristles on aluminium, although it occasionally needed a steel brush or point for stubborn or deep corrosion. The stuff that gets brushed off and even the wires from the brush can sometimes be a problem if they go where they shouldn't.

A lot of old component leads get some level of corrosion on them for various reasons, but the extra handling/cleaning/chemicals/soldering/etc. to remove it can easily result in 'collateral damage'. It can be a bit of a judgement call what to do.

Brushes like a toothbrush and small artists type brushes can be very good at cleaning around components even with basic cleaners like IPA, or water with a small amount of washing up liquid in it. A lot of electrolyte gunge dissolves better in water than alcohol.

[5065AGuru]

Maat,

The schematic you are using for the lamp assy is an old one.
The 130pf cap is correct. No tuning is done.
Tubular ceramics are 3.9 and 11 Pf and I've never seen one go bad.
The tiny cap is a .001uf monolithic radial.
CR1 is a multi-pellet diode and can't be checked with an Ohmmeter. (Never seen a bad one!)
Make sure when you remount it that you hold the transistor heat sink while tightening so as not to twist the transistor leads!
Also once you reassemble (you should have had two big nuts and a fiber insulator on the central lamp post) position the inner nut so that when you tighten the outer nut onto the fiber washer it clamps the outer magnetic shield between them as the outer shield forms a heat sink for t e lamp assy.!
How does the reflector look?

Cheers,

Corby

[maat]

@FriendLogic:  Thanks for the input. It's always good to get some second opinion on stuff. I am not too keen on using brass with aluminium, though. It might be a non-issue, but I don't want tiny pieces of brass to be the cause of galvanic corrosion. I know stainless is not perfect either, but better non the less. I also happen to have a stainless steel wire wheel laying around. So I will give that a whirl (pun intended). Component leads I usually clean up with a fibreglass pen and some IPA.

@Corby: Thank you very much for reminding my about the refractor. I took that apart as well (or at least I tried). I used a screwdriver with some tape on the tip and bit of gentle force so separate the refractor from the protector and it came off nicely.



But then the joy vanished. HP used silicone to glue the glass in place! I *could* try soaking it in WD-40, make a Peek stencil and use a puller to force it out...But I would say that refractor looks good enough .



Just some ramblings of a laser physicist: When/if I ever get to replace that window with an optical filter I will revisit that issue. But before I do that I got to talk to the other physics group across the hallway, because they have one of those fancy Anritsu OSAs. That might give a clue as to what to remove from the spectrum. But then again I might just as well rip out the lamp. Install a fibre pigtail, grab a 780 nm ECDL from the bench and drive that D2 line properly...I mean it is all there. The whole physics package. That line is big as barn…Driving from F=2 to F’=2 is fairly simple. Then repumping the decaying atoms from the F=1 back to F=2 would result in the desired dip in the laser intensity. Alright, enough dreaming!