Test Equipment Anonymous (TEA) group therapy thread

[Vince]

Wonder how easy it is to service those boards behind the range/selector switches, looks tricky from your pictures.

David

Geez you made me open it up again to have a closer look at how it's put together... you picked my curiosity.

Good news, see pictures : this function/range/attenuator block actually comes off in 2 minutes, fingers in the nose !
Just 2 screws at the front, 3 at the back, all easy to remove, and the entire block comes off, no strings attached... it just plugs into the back plane, simple as that ! 



It comes with the decimal point indicator neon bulbs attached to it because.. well it make sense, these bulbs are driven directly by the range switch.



Then, this block itself, is made of a backplane onto which the 6 boards plug into ! 
Of course they are all traversed by the shafts of the function and range switches, so I guess you would have to unplug the whole stack at once, then pull/slide them off of the shafts.

It does have some similarities to the A1466 in circuit design & layout, at least for the display counter boards.

Yes, though a more direct comparison could be made with my pair of A.1335 DVM, seeing as they too are monolithic entry level compact models.
As you said the counting par / display board, look identical, albeit with a slightly different layout to suit the size and dimensions of the boards, which are different. 
Then the most important similarity is the board that drives the symbol tube : here too it looks like on the A.1135 : that board does all the grunt work : takes a normalized input voltage from the attenuator section, then does the voltage/duration conversion, and feeds pulses to the counter/display boards.

So since I have studied this board quite extensively on the A.1335 (though I still have not managed to fix them  ), I am in known territory... so even without schematic I can at least do some basic troubleshooting.

I got carried away and removed all the Nixie boards as well...



Below a pic showing the symbol tube / "smart" board on the left, and on the right one of the counter boards.
The topology is just the same as the A.1335... only slightly more modern. Some of the film caps look more modern, I noticed, but other than that...




I also spent some time to test the zero offset problem more seriously, in a controlled way, methodically.... here are the results.



Volts and Amps ranges : left the inputs floating at first, then tried shorting the inputs... made zero difference whatsoever.
Ohms mode : I shorted the input, so at least I know it ought to display zero...
Capacitance mode : inputs floating with no test leads in the jacks, to minimize parasitic capacitance.

Because I have also unstable readings a sa secondary problem, I spent one minute on each range, the time for me to see a good sample of what readings it spits out, and take note of the min and max values, which I indicated in the spreadsheet.

Now.. I have some solid data to work from.. and I am not sure I like what I see. I get mixed/confusing signals here 

I mean, if you look at the Vac, mAac and mAdc modes : they all show an offset, which is constant no matter what range you use. This means the problem is common to all ranges, hence it would incriminate the "smart"/symbol tube board.... but OTOH in the ohms and capacitance modes, there is no offset, it works fine.... this does not add up. The smart board can't know what mode you are in, all it sees is a voltage, it does not know what range of mode it corresponds to...

Then worse : in Vdc mode, it's inconsistent, with 2 ranges showing an offset, and the other 3 ranges that don't ! 

So that means the problem might be in the mode/range block/sub-assembly indeed... which suck big time of course, won't be fun chasing / probing stuff in there...    The lack of schematic to teach me about this block, does not help...

That's all for now. You managed to make me stay up working in the lab 'til midnight, on a week day...

 

[Vince]

OK follow-up from last night.

I took apart the attenuator block..... as I hoped/imagined it's indeed freaking easy.
Just remove 4 screws to free the backplane board from the chassis :



...lift & flip (it's attached to the chassis by a wiring harness), then pull / slide the boards out one by one ! 



See something unusual in this pic ?



There are 6 boards, each with two wafers (well the last board only one), so 11 wafers in all... in a cramped space... it should be a huge mess of wires linking all the wafer contacts... but no ! Look not a wire in sight, where are they ! 

Turns out the contacts of the wafers are soldered to the boards ! So no wires needed, using PCB traces instead ! 
Never saw that in my old TE, first time.
How clean that looks, beautiful. That sure helped them fit all these boards into such a tight enclosure, as well as save an awful lot on labour in order to achieve the price point of this very entry level instrument. Win-win 



I took a pic of the first 3 boards that came out. The last 3 would have come out just as easily if it were not for a single wire connecting the red/positive input jack to the board that's right behind the front panel, then that board itself has one wire connecting it to the next-next board, though that is dead easy to get to... i's right on the edge of the boards, waiting for you.
Nothing too difficult to desolder, but sorry I was not going to do that for the purpose of this post... 
The board that's in between those two, is an EMI shield, just a 100% copper fill with no traces or components.



OK so I now put it all back together, and by some miracle it still works, no better or worse than before I took it apart. The clever construction of this thing helped a lot with that...as well as gold plated contacts everywhere.


This attenuator block is actually even easier to remove than I said last night : it's not 5 easy screws but only 3 ! The two screws at the front didn't have to come off. They hold the bracket onto which the neon bulbs for the decimal points are mounted. I freed it assuming otherwise it would catch on the display boards... but no, it does not, so you can leave it alone still attached to the attenuator block... and the better for it because putting these two little screws back, proved to be challenging, took me 10 minutes....I hate slotted heads...

Noticed that the orange colour filter for the display, was cut 3mm too short or so, shame, so it can be that if you don't pay attention, it can slide around and you see a gap between the filter and the bezel inner edge    So I made sure it was adjusted as best it can.
That filter is actually just a thin/flexible plastic film, it's not a rigid thick piece like on my other Rochar TE. This film rests onto a rigid transparent piece of plastic. So I guess one improvement would be to source a coloured plastic film and cut a piece that's actually the proper length...

The bezel turns out, is not a separate piece, it can't be taken off the front panel ! Nope, it's just part of the Bakelite enclosure, it's one big, single piece of Bakelite.  Cost cutting again. They really scratched their heads to find ways to lower the production cost of this unit...

[tautech]

Vince, HP1740 scopes use a similar stacked PCB arrangement with a shaft running through at least 3 PCB's that I recall. However it was a double shaft, one inside the other and need be timed correctly to remove or reinsert the shaft.
Sure don't miss them days. 

[factory]

The A1613 looks very nicely designed, thanks for all the extra pictures  , guess you just need to something better than the reverse engineered diagrams that are on the web.

David

[tggzzz]

Vince, HP1740 scopes use a similar stacked PCB arrangement with a shaft running through at least 3 PCB's that I recall. However it was a double shaft, one inside the other and need be timed correctly to remove or reinsert the shaft.
Sure don't miss them days. 

Yes, with an extra twist...

The switch contacts are the usual gold-plated fingers, reliable but might need occasional cleaning.

The rotating switch housing contacts the PCB, and eventually abrades the PCB track leading to the contacts. Bad design, but repairable.

The 1740s do have a delightful failure mode... After a while the risetime degrades significantly.

The cure is to caress, squeeze, and generally fondle the coax delay line for a minute or so. That restores the risetime to 3.5ns. Seriously!

[factory]

Oh 5340 day! Did I miss it?....

@Factory, I probably have some parts if you need any to complete yours. I'm certain to have those two missing cards in the aluminium box, and maybe a HP logo too. 

Thanks for the offer, but I got lucky on ePay a while back, someone bought a functioning 5340A at a HAMfest and then promptly destroyed it  :'(. All the small boards got offered as a job lot (which I won at a very good price) and the display board separately (didn't win), I did ask about the rest, but got told they had gave the frame & other boards to someone else.

I've actually got four of these things here now, the early one (with it's strange modifications), one fully working one, the mostly repaired one from PP auction (still need to get the transformer rewound) and finally an LED version which is untested (sold as working) but needs the RIFA madness smoke bombs defusing. 

Extra info, I've since found out this is another HP counter that uses regular 7441 display decoders, early one uses 1820-0426 (7441), later one uses 1820-1161 (HP clone of 7441?? I need to confirm), display blanking is done with separate trannystors.
I also found out they replaced the 1820-0426 with 1820-0707 (74141) in the 3450A/B DVM, still got to find one of those DVMs in the UK/EU.

David

[Vince]

The A1613 looks very nicely designed, thanks for all the extra pictures  , guess you just need to something better than the reverse engineered diagrams that are on the web.

David

Yes, my kingdom for a service manual........

[Carl_Smith]

Does Shahriar have enough test equipment?

[AVGresponding]

[zrq]

Does Shahriar have enough test equipment?

Did he had videos featuring the Thorlabs piezo drivers and the New Focus power meter?

[m k]

The A1613 looks very nicely designed, thanks for all the extra pictures  , guess you just need to something better than the reverse engineered diagrams that are on the web.

David

Yes, my kingdom for a service manual........

Google seems to be extremely limited these days.
Those were the days when Altavista just puked everything out.

Have you asked AI?

[m k]

..
Berești-Bistrița is a county in Romania.

Actually it's more like a small township, but how did that come about?

I'd say that you didn't really miss anything.

And it seems that Romania has a different governmental structure or naming system.
There county is actually quite big entity.

[Vince]

The A1613 looks very nicely designed, thanks for all the extra pictures  , guess you just need to something better than the reverse engineered diagrams that are on the web.

David

Yes, my kingdom for a service manual........

Google seems to be extremely limited these days.

Well it can't show me something if it does not exist....

Those were the days when Altavista just puked everything out.

Altavista was not that good. I remember 25 years ago asking it what were the winning numbers for the next national lottery, and it failed miserably.

Have you asked AI?

Who's Al ? What's his line of business ?

[TERRA Operative]

He sells shoes I think.

[Vince]

Looks like Al won't be much help then...

[BU508A]

Who's Al ? What's his line of business ?

I guess, he meant A.I. = Artificial Intelligence

[Vince]

Of course he meant A.I. ... I was just trying to have a bit of fun because Al and AI look the same with most fonts... sorry 

Took me a few seconds to realise it was you writing... I see you changed your avatar picture....

[Vince]

Working a bit on my Metrix 920 RF gen this afternoon.

Thinking of selling it to downsize.

Bought it 15 months ago, back then fixed a failed filter cap in the PSU and then it worked fine.

But of course I wanted to try it again to make sure it still worked, because I am that nice.

Well I am glad I did that because for sure it didn't work any more !
Looks like sitting for a year in heated house does damage...

1) For the life of me I can't put my hands on the power cord ! Uses an obsolete plug of course, so it was vital 
So had to resort to a sketchy montage to power it up. Don't try this at home.

2) Had no output at all... eventually found out that inside the attenuator assembly (that big aluminium cylinder in the inside pic), the wire that goes to the center pin of the coax connector on the front panel... somehow came off. Solder joint broke.
So I soldered it back but... after only a couple seconds of me heating up the pin, said pin would move around ! Yes, the dielectric in the coax connector was softening so bad, like butter !   Yes, I was using a soldering iron not an oxyacetylene torch...
Could not believe my eyes. I guess that's the final nail in the coffin for this connector, which has been the subject of a much heated debate here on TEA at the time, I remember vividly...
I could not buy anywhere a BNC adapter that would fit. I could find an adapter, but the belling-lee side of it was not split/slotted, so it had no springiness to it, keeping it from mating with the socket. So Specmaster suggested to replace the connector with a panel mount BNC connector... tempted now, or not. If I am selling it, no need to bother.

Anyway, this fixed it ... or part of it. I now get a signal on the front panel.. but only the LF / 400Hz modulation frequency.

2) HF doesn't work, yes two problems at once, my luck. 

There are only two tubes in this thing (other than rectifier tube). Both are 6J6, dual triode.
One of them makes the oscillator for the HF. The other has one triode doing the LF oscillator, and the other triode mixes the LF and HF to produce the final signal that goes to the front panel.
So I swapped the two tubes, and hey presto I now have HF ! 
So I have a dead 6J6 tube.
10 Euros NOS warranted from my local shop. Was already planning on ordering a couple things from them in the coming days, so I can just add that to the list.

3) A year back this instrument was starting to try to zap me when I was touching the case.
The Y caps were failing I guess, classical.
Well today they failed for good. After a few minutes working on the thing, the differential breaker tripped.
So I removed those caps, and I am back in business.

So.... a bit of a tough ride.

I think I will invest 10 Euros to replace the tube and have a working unit, will help in selling it I guess, and balance the fact that it's now lacking its exotic/obsolete power cord.  It's in great condition cosmetically so it could appeal to "deco" fans as well, quite trendy these days over here.
Hopefully one day I will eventually stumble upon that bloody cord....

Might fit a modern day earthed IEC  socket at the back, for safety, but not sure it would appeal to vintage TE collectors who might see this as mutilation.
Same goes for fitting a BNC connector at the front.

So, first things first.... let's buy that 6J6 tube to bring that TE to operational status, then we will decide what to do. Hell I may even be glad to have it working again and decide to keep it... after all its goes up to 50MHz which is quite useful, and it's not a boat anchor, not that large nor very heavy.... hmmm... yeah I am hesitant now    I like the looks of it the quality look and feel of the engraved stainless steel dials...

Time will tell... maybe I will flog other stuff first, and then come back later to it to make a decision...

At any rate these are hardly rare over here, Metrix being local, lots were sold and they are found routinely for sale here. But that one is in excellent nick inside out, and I paid only 30 Euros for it so... it I were to buy another later, it would be easy to do I think, but it would not be as good value for money I think....

[m k]

Well it can't show me something if it does not exist....

Yes, but doesn't it.

Google is clearly a bit protective when somebody starts poking its secret info stash.

[m k]

Of course he meant A.I. ...

Was it double, I can't tell.
Pretty good one if it was.

[Vince]

Next in line in my spring clean / downsizing is this Tektronix 453 scope.

50MHz, like new inside out, needs to be seen to be believed.
Comes with its cover which doubles as a storage compartment, has at least one period probe IIRC.
It's one of the "original" ones... I mean it's branded IBM. There is even some IBM yellow paper memorabilia in the storage compartment.

Sadly I bought it defective and it turns out it's not worth fixing. I lost my gamble. Lights up but no trace on the screen. Keeps blowing the HT fuse, and asI had it apart trying to troubleshoot the CRT HV section, I saw with my own eyes the HV transformer escape the magic smoke, which although I am no expert, can't be good.
Datasheet for the transformer is available, if you feel like winding your own transformers.. I don't.
Plus, the root cause of the problem might elsewhere and the transformer might only be a collateral damage.
I don't want to spend time nor money on this puppy, let's be honest.

It would be worth fixing if I were in love, but I am not.... I didn't want to buy it initially, but got persuaded by one of you, shall not give names. I fell for it eventually, because the seller was barely 10kms away from me, so no shipping cost involved...

So if you are a fan of these scopes, anf want to save its life, it's yours for the price of packing (20 Euros) and shipping via Mondial Relay (25/30 Euros to Germany IIRC),+ 10 Euros just to say I didn't give it away... (paid 40 Euros for it a year ago IIRC)

Can't sell it locally since a working one is not much more than the broken one I have... and the price of shipping and packing is a show stopper.

So the only thing that makes sense financially, is to take it apart and salvage a bucket load of like new parts that are worth way more than the non existing profit I could ever make from selling it, assuming anybody would volunteer to buy it.


So, In order not to repeat the mistake I made with the Tek 7504 I took apart the other day : are there any parts of particular interest in this scope, other than the obvious general purpose stuff (all the boards and switches/pots/trimmers/knobs etc) ?

The 4 feet at the back are all gone, disintegrated, as they always do on these scopes, I gather.
I think there must have been feet on the bottom as well, but they are all gone as well, only the metal work is left...

Maybe the trannystors for the vertical deflection ? They are mounted on the chassis. I guess they must have some good bandwidth.


So if you need some part to fix your scope, speak up....


@ Neomys : the current calibrator fixture ? I don't think it's the same model as the other Tek, so you might not be interested I guess.

[Neomys Sapiens]

@Vince: it does not even have that little bracket.

[Vince]

It's on the side of the scope.

[TERRA Operative]

Here's a question for the tool aficionados...

I need to tighten the 5-turn knob on a multi-turn pot on my Tek SG504 (same thing is used on other gear like the Tek SG503 etc) and it seems I need a special little socket wrench..

Does anyone know what this sort of thing is called and where I can get one? I swear I've seen them before but I can't get the right keywords to find them in google..
I need one with an OD of about 8mm or so.

[tautech]

A minute or 2 surfing the net identifies such drivers as SNAKE EYE security bit/tips.
https://www.anzor.co.nz/security-screws/security-tools/snake-eye-bit
Hope that helps find what you need.