Surplus Equipment, Not Optics

[mawyatt]

I’ve been adding some surplus equipment to "Mike’s Labs” lately. Now I’ve got a Tektronix 2465 an Agilent 34401A and HP 34401A for starters.

I needed a precision DVM for some analog precision work (16 and 20 bit DAC and ADC) since my 30 year old Fluke 87 is limited resolution and accuracy questionable, so I got a old HP 34401A off eBay. It arrive physically damaged in shipment so the accuracy was of concern. The seller offered me a $50 rebate to keep it which I did, but still didn’t know how good it was. After fixing the HP, the Fluke and HP agree reasonable well, so suspect the Fluke is good, but still didn’t know how accurate the HP was. I've looked into recalibration, then a reference standard…all way too expensive, even surplus (may design my own based upon a 0.05% voltage reference, and some precision resistors and op amps).

Since I don’t have access to any calibrated equipment anymore (retired now) I decided to get another 34401A and got a good deal on the Agilent version, which is newer and suspecting it's more accurate. Good news is the two agree within about 10uV at 5 volts, later I’ll check the resistance and current scales but they look really close as does the AC readings with just quick viewing.

Also needed a good analog scope (don’t like the digital scopes that much) and won a eBay bid for a questionable Tektronix 2465. It arrived in a horribly packed box, side split out and the arm was damaged. I decided to keep the scope since it was a good deal (seller wouldn’t offer any concessions, just a full refund for return) and the 2465 & that series of scopes are considered the finest analog scopes ever made, and I agree they are absolutely beautiful instruments.

I checked it out and things appeared to work OK, but some excess noise was on the traces but still usable. Took it apart and fixed the broken handle, then got out some Lysol, alcohol, acetone, Windex, vinegar and citrus cleaner. It was filthy and required serious cleaning, then used some Meguire’s Auto plastic restorer on the plastic, now it looks great!!

I noted some capacitors that are leaking and made some measurements. Many of the supply voltages had excessive noise and the -18V was really bad, these are all derived from direct line SMPS, so lots of switching noise. So decided to replace as many of the caps as I had available.

In the meantime I won another low bid for a Tek 2445. It arrived very well packed and upon power up and self test the screen locked up, then it went up in smoke!! I checked things out and the power supply was partially toasted, so just sent it back, but I’ll look for another 2400 series “bargain” scope that I can fix.

Here’s all the capacitors I replaced, have more on order and will replace them later. This cleaned up the trace and it passes all self test :>)

Must admit it's kinda fun with the older instruments and bringing them back to life And you save a bunch of $ too

Best,

[ray_parkhurst]

Shame about the 2445. Did you bring up the AC supply slowly to reform the capacitors? For anyone thinking about buying older equipment on eBay, it's not a good idea to put the full AC voltage on the equipment right out of the box. If the equipment has gone more than a year without being used (you must assume anything bought from eBay or surplus stores has gone unused longer than this) the electrolytic capacitors can become leaky (electrically), and need to be charged up slowly so the dielectrics can self-repair. When I do this I usually take the voltage up to the max ratings of the equipment, and maybe a bit beyond, to ensure the capacitors have reformed to the max voltage they will ever see in use.

[mawyatt]

Yes, shame about the 2445, it was in good physical condition. The failures were a couple, the screen trace locked up, then started displaying gibberish. I couldn't enter self test after that, then it started smoking!!

One of the infamous "RIFA" metalized plastic capacitors ruptured, this caused a 68 ohm 1/2w series resistor to completely burn up, causing the PCB to also burn.

Figured it wasn't worth the effort to try and revive this one, and I already had the superb 2465 working, so no urgent need.

Agree, probably a good idea to slowly increase AC line, rather than just jolt it ON initially after a long period of being stored. I'll need to get a Variac tho.

Here's the 2445 screen and power supply. Note the replacement of the bridge rectifier with discrete diodes! Seems this power supply has had issues before!

Best,

[mawyatt]

Another Tektronix 2465 to see if it can be revived. Built-in self calibration waveform is all hosed up, and both input channels 1 & 2 aren't working properly.

After quite a few hours of disassembly, diagnostics and measurements over the past few days the input channel issues traced to the input alumina substrate hybrids, which are not repairable and must be replaced as matched pair according to Tektronix service manual.

Self calibration problem traced to the waveform creation circuity.

Was sold as a working unit, question now becomes should I keep this or return it

Best,

Top 2465 has the correct waveform.



Calibrator waveform, should be square wave of 400mv peak amplitude like above.


Large main circuit board with various hybrids, including the suspect input hybrids.

Suspect input hybrid.

Disassembled input hybrid.

[mawyatt]

We decided to keep the scope and try and fix it. Seller made a reasonable partial refund since we didn't want to try gouge an them, they aren't engineers so really no fault on their part not knowing about these detailed issues. However, the scope would be next to worthless without the input problems corrected.

After a few more days of troubleshooting we found the issue with the calibrator and input channels 1 & 2. The calibrator had a defective custom Motorola zener diode which was replaced with a 1N4733A after a little circuit analysis to figure out what value it should be. This is a superb little design in that it produces a very accurate zero to 400mv square wave output, so the DC value is 400mv/2 or 200mv and the AC RMS value is 200mv. RMS is Root of Mean Squared, so square root(average((+-200mv)^2) is 200mv! Brilliant little design concept, and can be checked with a good DVM to verify.

The input channel 1 & 2 problems are with the hybrids as mentioned and these are no longer produced. So a donor scope must be found and the input hybrids surplused. We decided to try and repair these since they were useless anyway. After some more troubleshoot an open solder joint was discovered on one hybrid and since the input ground switching wasn't working suspected the open frame relays which use the alumina substrate as contacts. A new rubber pencil eraser was carefully used to resurface the gold contact planes, then thoroughly cleaned without a solvent. Dark residue was observed on the eraser indicating surface contamination, probably tobacco smoke.

Resoldering the hybrids proved very difficult and care must be observed to not leach the hybrid delicate surfaces. After a few assembly/disassemebly cycles where soldering the hybrids back into the main PCB cause the hybrid interior contact to de solder, we had to set the iron temperature to 200C, just above the melting point of solder, this worked.

After the hybrids were installed, everything went back together, only to find what appeared to be a totally trashed scope, nothing worked, the display horrible and way too intense, no working controls, no readable text or waveforms, couldn't even enter the self-diagonsitcs. Thought we had completely trashed a beautiful scope

Another disassembly, probing reveled that one of the many (~20) voltages was not 42V but 3V. The power supply had the correct value but not the main PCB, suspect 10 wire connector proved the fault. This was a soldered-in connector that was replaced with a direct wired solder to the PCB, actually looks and works better than the original soldered in connector

Now reassembly for the 6th time, and the scope works!! Success finally.

Was it all worth the effort and frustration, probably not to anyone that doesn't appreciate these old instruments, don't think analog scopes are made by anyone today, so a true antique!!

So now we have 2 HP/Aligent 34401A 6 1/2 digit DVMs and 2 Tektronix 2465 scopes, all for less than the price of a new good digital scope

Best,

[chris_ma]

Hi Mike,

thanks for sharing these experiments, I very much enjoy reading your reports even though I understand only the basics (if that), but I can still somehow relate to it and get an idea how to tackle these things.

I'm also glad that somebody is repairing these objects since I'm afraid in a few years less and less people will have the necessary skills
chris

[mawyatt]

Hi Mike,

thanks for sharing these experiments, I very much enjoy reading your reports even though I understand only the basics (if that), but I can still somehow relate to it and get an idea how to tackle these things.

I'm also glad that somebody is repairing these objects since I'm afraid in a few years less and less people will have the necessary skills
chris

Hi Chris,

Thanks for the reply. Not sure anyone would even care about this kind of stuff here. Glad to see someone interested.

These old instruments were fundamentally designed and built to be maintained and repaired by others than the OEM factory. Don't think that's true today with anything electronic, mostly is if it breaks throw away and get another. However, today's electronics are way more reliable than way back, so things generally don't fail as often. Think this is mainly attributed to the higher levels of integrated circuits.

My time isn't valuable anymore since I've retired, so spending loads of time of fixing these instruments is worthwhile even in just saving an instrument from the junk pile, kinda like restoring an old car

Best,

[Troels]

Hi Mike,

I also enjoyed your story of trying to save old pieces of quality equipment.

I have also had great fun trying to rescue an old 24" HP plotter/printer from the trash fill. A friend got it for free with ink cartridges and several 24" rolls of paper because it was not working any more and considered outdated.

Well, it was indeed 25-30 years old. But loking at the construction you would think it was made to last forever.

I succeeded in finding a service and repair manual on the net and disassembled the machine. A rubber timing belt was worn out and broken leaving behind a lot of black, sticking dust.

I found a HP enthusiast in USA still selling the original timing belts and managed to clean everything and get it running perfect again. It was really a joy to reassemble all those ingeniously designed pieces and get them back again in place. It was designed to be serviced for ever.

The only really annoying thing was these ink cartridge heads always drying up. After lots of shaking and washing I was able to make a few big prints of my photos. But the costs of buying new cartridges and growing difficulties with getting the discontinued models as well made me give up at last. I think this machine was alaso designed to be used at least once a week.

But at least it has now been responsibly disassembled into plastic, metal and electronics, to make reuse of materials easier. Rest in pieces.

[mawyatt]

Hi Troels,

Sounds very interesting. Did you consider refilling the ink cartridges?

Best,

[Troels]

Not really.
Besause the HP cartridges are complex units consisting of both the ink tank, some electronic curcuits and a regular print head with nozzles for ejecting the ink.

Every time you change the cartridge you get a new print head with clean nozzles. If you don't use the printer for some time the ink in the nozzles dry out and blocks the passage of ink.

The printer has of course an elaborated system for protecting and cleaning the heads. Because this system used several different types of rubber sealings and plastic/silicon tubes it was too difficult for me to repair to normal working state. I could not prevent the cartridges from drying out.

Perhaps I could have solved this problem too with more investments in spare parts and working time. In the end I gave up partly because the printing quality was not up to modern standards. This machine was designed as a color plotter ment for techical drawing and engineering, not for printing photographic art.

[mawyatt]

I remember reading long ago about the amazing effort that went into designing the original ink jet printers, they studied all the properties of the droplets configuration, velocity, shape, viscosity and so on to get the best possible deposit on the paper. The ink jet heads are a very clever device, originally thermal, then later some used piezo electrics devices.

When I used my large Canon photo quality printer a lot, we refilled the cartridges and "fooled" the printer into thinking this was a new ink cartridge by resetting the cartridge chip. I don't print much anymore, only use a cheap HP, the Canon is packed away and the heads are probably dried out.

Best,