Test Equipment Anonymous (TEA) group therapy thread

[Vince]

QUAD 33/303 Amp restoration

Worked a bit on it this afternoon.

I finished restuffing the old can caps, and fitting them to the amp. I am happy with the result, though of course this first hand-on experience was educational and will allow me to do a better job on the next piece of TE.

It's a success as far as I am concerned. I get to conserve the original look, conserve the original brackets so that the caps are perfectly secured, and conserve the original cap terminals which allowed me to redo the wiring for something acceptable, almost decent, replacing the true butchery that I had to resort to initially when I tried to solder all those wires directly to the terminals of the new snap-in caps.
One of the terminals has no less than 4 wires going to it, one solid core wire and 3 stranded wires. Try soldering that to snap-in terminal, and make it look good...  for this alone I am so glad I have not the original terminal to work with. Still was not fun to do, but at least it was possible to do, given enough care and patience...

So then I powered it up and let it cook for 5 minutes, to let my new caps a chance to explode, in case I mistakenly wired one or more of them backwards. I was super careful throughout the restuffing and rewiring process of course but you never know. There is a lot of energy in those 4 big caps, so safety first...

Then I hooked my sig gen and a pair of speakers I salvaged a couple weeks ago from an old flat TV I was given.

It worked just fine for a few minutes, then all of a sudden the right channel quit ?! I was about to throw this thing out the window (again) and flog it for parts on ebay for 50 Euros 

Pulled the plug, flipped it to see the solder side of the power boards... and immediately found the problem.. see picture.



The two mid power trannies that drive the final drivers (the TO3 mounted on the rear heat sink) ... they looked weird : their pins were sticking out the PCB surface as if I had forgotten to cut the pins after soldering them... strange. Looking closer... all the tracks leading to them, were broken ! 

So my only guess is... these trannies overheated badly, somehow enough to soften the solder joints sufficiently for the trannies to "sink" through the PCB... and break the traces on their way.

The original trannies were small TO39 with a built-in small metal plate / heatsink, mounted on a small flimsy heat sink as well.



They were dead so I replaced them with what I had on hand, temporarily, for diagnostics purposes, with TIP31/32 IIRC. Worked just fine and with their beefy TO220 package, was safe enough I thought, not to use a heatsink if I kept the output power/volume low.  Well I guess they do need a heatsink anyway then... 

Well, I hope that is the actual problem, and not the symptom of a deeper problem....
So for now I will do as if it were the actual problem, and just wait for the replacement trannies to arrive from Mouser.

Well I have the 2N5320 "proper" replacement trannies here (well, what the QUAD people use to replace the old unobtainium trannies anyway), for a while now, because my local shop had them in stock, but the 2N5322 complement, they were out of stock 6 months ago and still are today.
But Factory ordered them in his latest Mouser order, so they will arrive eventually...

In the meantime maybe I could start work on the 2N5320 I do have, which are TO3 but with no metal plate, to figure out how to reuse the metal base of the defunct original tranny, and fit that to the T03 package. My soldering iron is only 50 Watts so I need to experiment to see if it's at all possible to solder the plate to the T03 package.
If not.. maybe I could instead glue the two parts using thermal glue, the stuff that's used to glue small heatsink to flat IC packages... maybe that could work here.

Anyway, need to experiment...



Also... as far as the 33 pre-amp is concerned now, there is progress too. It was missing a front tooth, poor little bugger... hard to sell like this.
So I finally fitted the missing knob, looks proper now !
I am so happy to have that fixed, as this button was so hard to find... literally only one on the planet. Thanks factory for the assistance on that... it was worth the trouble and silly cost



OK so after a year working on this amp duo, I am now starting to see the light at the end of the tunnel.... 

[m k]

Back when I was a sparky in Australia I would sometimes come across the single-pair version of that cable in older houses.

Still worked fine, and I guess the outer sheath provided pretty good shielding from interference, and gamma radiation.

Oh yes. Our house is built in 1930 and previously my VDSL2 line on a single pair, lead shielded cable gave me 100Mbits/s down. I admit I was quite flabbergasted at the time.

Wilko

My previous home was built in the 70's and had aluminium telephone cables. They struggled to do 56k 

Our ADSL "copper" pair from same era was once sort of unstable.
One wire was cut.

[m k]

getting polarisers for FSTN or STN screens is proving difficult

Have you tried atypical sources?
Glass industry has all kind of films.

[Vince]

QUAD 303 Vintage amp restoration

Hmmm...

This evening while everybody on earth is watching their favorite american series on TV or streaming.... I am working on trannystors.

I had a first go at "recycling" the metal plate of the old transistors I need to replace.

Decided to first drill out the guts of the old transistor from the bottom, using the thick plate to help guide the drill bit.
Measured the inner diameter... about 7.3mm.... Believe it or not I DO have a 7.3mm bit !
Not even kidding... when I left my old work place, one of the workers gave me a random selection of drill bits as a goodbye. They are all of odd sizes because they are sizes that we use at work heavily for common rivet sizes, drain holes in stringers, locating holes in frame segment, that kind of things.... and I found a 7.3 !!! 

So I was able to drill the guts out :



..but the can itself was still in place, it survived.... so I chopped it off the plate with a hack saw, then filed the plate flat to make it nice and smooth.




Then I picked some random TO39 transistor in my pile of boards, and tried to see if I could solder it using my under powered 50W iron...
At first I thought no way, and was not even sure if the plate would take solder at all to begin with...
Well, it was a bit of a struggle for my iron, but it CAN solder the can to that plate !!!!   

I used my finest solder wire, for SMD word, 0.4mm diameter, to try to make a seam as thin as possible... it worked quite well.
Looks horrible on the pics, but in the flesh it looks quite nice especially for a first attempt with an under powered iron.



I could file the seam down to make it look prettier. Might do that.

Anyway, that's cool.... I now know it CAN be done with the gear I have ! 

Now the next problem is.... how can I do to hold the can in place while soldering it to the plate ? It needs to be perfectly centered and with the proper orientation of the pins of course. Here I paid no attention to that of course... aim was only to validate the soldering process on its own.

Any idea ? Maybe I could put a couple dots of varnish or superglue to hold the can in place, then I could put it in the vice like I just did, and solder it.

[factory]

Looks quite good actually, I didn't know you could solder those cans.

Could you use a piece of circuit board as a jig?

P.S. TV hasn't been on this evening, been busy recycling some modern electrical crap, two fan heaters, bosch strimmer and a twin halogen light thing.

David

[Vince]

PCB  of course ! Stupid me....

All I need to do is temporarily fit the plate and T039 tranny onto the driver board, minus the heatsink of course, this way both the plate and can will be in their perfect location and orientation and secured properly !
Then I can solder the can in situ, then remove the now one piece transistor, fit the heat sink, solder it back to the board, et voilà ! 

To make it easier for my under powered iron, I think I could try preheating the plate with my hot air station prior to screwing it down to the PCB, or maybe hover it for few seconds over the gas burner in the kitchen.

I really need a more powerful iron... think the next size up is 90W then 130W or something... want that. With a big chunky tip / high thermal mass.

[factory]

I actually meant making a jig from a piece of spare/scrap PCB and drilling the holes to match, plus it would be easier to access for soldering the plate & trannystor together, less risk of damage to the old PCB.

David

[Vince]

This is the biggest piece that I brought back from my big TE hakl a few months back.

This DOES qualify as a boat anchor.... it's huge, it's cubical, and it's freaking heavy.
I am big guy and am very much used to my 30kg old Tek scopes... and this TE is significantly large and heavier than my Teks...

It's so big and heavy that at first I refused to take with me... but then when I had a closer look at it, I saw how cool it looked, oozes build quality inside out, show room condition inside, excellent nick outside, and  a general purpose enough and with decent specs, that I could potentially use it.

So I didn't have the heart to leave it there, know the guy would scrap it soon.

It's still too big and heavy for me to contemplate using it.... but it's still nice enough that I am not in a hurry to get rid of it. I will take a few months to try to find it a good loving home.

In the mean time I am very excited and curious to see if I can get some life out of it, and see if I can get it working !
Would also help selling it as well...

So what is it then ?

It's a Ferisol P201A pulse generator. So it's French stuff, and was owned by the French army according to the plaque riveted to its side.



I can't find a service manual for this thing... if you do please share.

From the front panel, we can at least see the features and basic specs of the thing.

- Pulse width adjustable down to 50ns.
- Trigger : external, manual or via a built-in oscillator adjustable up to 100kHz
- Output : 50 or 300 ohms, positive or negative pulses.
- Amplitude : the attenuator goes from 0 to -59dB



Construction

Like all decent boat anchors, it's got loads of tubes, cool dials and big knobs...



Nice mechanical construction of the attenuator. The knob is two-part : the black part makes a window slide up and down the dial, to select the coarse rang : 0-9db, 10-19db, 20-29dB etc, up to 50-59dB. Then if you turn the red knob, it makes the drum of the dial rotate which lets you fine tune the attenuation in 1dB steps.



Each knob rotates a camshaft via a bike chain. the camshafts have 4 cams and the lobes actuate 4 switches in a binary pattern. The switches of course take the relevant resistors in and out of action.




Other than this attenuator, there are 3 more dials on the front panel, which are quite nice : they have an anti-parallax system. I tried to show it in this picture, not easy.
You have a very thin fixed needle / marker... that's doubled : there are two needles, a millimeter or two apart. So you line them up tio make sure you are reading perpendicular to the dial.
Attention to detail.



At the back there is a convenient mains voltage selecting switch, hidden behind a small door easily popped, secured by two quarter turn screws. So first thing I did was set it to 240V....




The instrument comes apart dead easily : put it face down, resting on the two beefy bar handles, which conveniently protect all the knobs... then at the back there are 4 giant thumb screws.. remove them, the cabinet just lifts up in seconds, and you are in ! 



Electrical testing

Since the instrument passed the visual inspection with flying colours, like new, no damaged or missing parts, all knobs that rotates freely and feel smooth...
I could switch to the next phase : checked fuse : good.
Grab DMM, test the power socket : primary of the transformer is neither open-circuit nor shorted : reads 3 to 5 ohms which makes sens for a transformer that... gigantic.
open-circuit between ground and neutral/live pins. Checked visually : can't see any Y caps anyway. So I am safe here (unlike the Metrix RF gen that tried to kill me with leaking Y caps and is not even earthed...).

Spotted to big can caps. They were neither shorted nor open-circuit, read a few megs, so looking good. Measured at 200+ µF and.. they are marked 200µF so.. I am happy.

So with these basics checks done, I was OK powering it up... through my dim bulb tester. It's got a small 77W halogen bulb so of course I knew it would light up solid which it did, but the point was not to get the TE functionnal/working, but just to trickle a few mA at a low voltage, to wake the TE up slowly, let the caps get out of hibernation slowly... and limit the magnitude of massive component failures, should any happen.



I measured the AC voltage at the power pins.. got about 25V.
then measured across the two can caps I had just tester : got 30Vdc with negligible ripple, so was happy.
I watched the voltage, noticed it would climb, slowly but steadily, mV by mV.... I monitored it for 10 to 15 minutes. Eventually it capped at 40 Volts or so, at which point the massive beefy chunky scary cooling fan.... came to life  Yes, it started turning... barely crawling but still, there was life in this TE.
I left it warming up for another 30 minutes or so, voltage is stable, no explosions or smoke.... so me thinks the caps are good enough, and now it's only the bulb tester that's keeping it from coming to life. So I feel confident enough that I will now take the risk of powering the thing directly from mains, no dim bulb tester anymore.... will that prove to be a mistake ? Will it come to life or explode or smoke ?

You shall have the result in a coming post... for now I only wanted to introduce this TE, wet your appetite, and use the time it took me to write this down, to give the TE some more warming up time.

Oh forgot, I spotted a convenient test port in the PSU section. A connector that brings you all the power rails, clearly labeled, easy to reach, with convenient 4mm jack connectors to plug your DMM. All ports read dead zero. Of course.



OK so let's upload some pics then try to power this thing up and report... I am so excited !!! 

[PA0PBZ]

No service manual but I found this for you, in your 2 favorite languages 


Edit: There's something strange with your last post, clicking a picture makes it disappear and clicking the link says I'm not allowed to that section? 

Edit 2: Looks like the post is gone now...

[Vince]

Oh fuck me the forum is screwing up big time with image attachments and thumbnails, even though I skipped the first one of course.

Please bear with me as I keep uploading the same pics over and over again hoping one day it will actually work properly !!!

[Vince]

The rest of the pics.

[Vince]

No service manual but I found this for you, in your 2 favorite languages 


Edit: There's something strange with your last post, clicking a picture makes it disappear and clicking the link says I'm not allowed to that section? 

Edit 2: Looks like the post is gone now...

Thanks for that document ! 

Tells us something new : rise time is 20ns !

Hopefully the instrument will work and we will be able to measure that 


I have finished my long post, inserted all the pics.... to my end at least, it all looks to be working. may it continue this way for more than 5 minutes ! 
 

[Vince]

hmmm... refreshed this page and got an error page, saying the "SMF server was having difficulties accessing the database"... might explain my picture problem I guess...

[TERRA Operative]

I see lots of pictures, so something got through. I am seeing the SMF errors too, maybe the forum server needs a bit of a slap around... 



Tonight it was Dallas NVRAM time. I cut the top of a DS1245Y off to remove the old batteries and chiselled away at the corners to expose the terminals. Then some superglue had new battery holders fixed in place with a few jumper wires added to connect things together.
Tomorrow I'll get some batteries and see if this thing works....
I ripped the contents already, so once the new batteries are installed I'll copy it back, pull the chip out of the programmer for a few minutes then stick it back to see if it's all still there.


Then it goes back into the TDS540. I ain't spending fifty bucks on a new one if I can get it working for under five... 

[Vince]

FERISOL P201A

OK so I fired it up !!! 

That's all, have a good day.  Oh no, you mean you want more details ?!  OK fasten your seat belt then...

The thing powered up, no explosion. The fan went full speed, good.
However a few just a few seconds, things went very wrong : a horrible, very loud buzzing sound tried to make my ear drums explode.
A handful of seconds later, noise stopped and so did the instrument... fan stopped, everytning stopped.... the thing blew its fuse !
It all happened so quicly, I did't have time to understand what was that buzzing sound, nor locate it. It reallt took me by surprise as I was expecting fireworks or smoke or caps exploding, but my brain was not prepared for a loud buzzing sound !

This TE is full of surprises....

My best guess was that it must have been some relay contact going nuts, vibrating at high frequency...
So I thought well maybe it's a relay in the PSU section to switch the outputs when the thing is warmed up...like it exists on some Tek hollow state scopes.
So I searched for a relay in that area and I found one !  See pics.
It's  right next to the transformer and the two can caps I checked.



There are loads of pins on its socket, but only a few are wired. Two thin wires on the RHS, the coil most likely, and on the LHS 3 beefier wires buggering off to the transformer it looks like...




So I did the basic checks again : transformer primary still not shorted, still reading 3 ohms or so. Still no short between earth and live/neutral. The can caps are still good hmmm.... so other than the fuse blowing, it all looks fine, the fondamuntals are still good ?

So I replaced the fuse, put the dim bulb tester back, flicked the switch and... it behaves 100% like it did before. Starts at 30V, slowly climbs at 40V then the fan starts rotating slowly. Hmmm... so no harm then, back to the beginning ?

So I tried my luck and again powered it up with no dim bulb tester. Of course this time I was prepared, so I would not be scare of the buzzing sound, and would rather be ready to do look around searching for clues.

And the exact same thing happened : it does power up, tubes light up, frnt panel lights up, fan speeds up, all looks good, then again after 5 r 10 seconds it buzzes at me. Except this time I pulled the plug immediately, not waiting for the buzzing sound sto stop, in a bid to save the fuse.... and it worked ! I saved the fuse   

So in the 5 seconds that this spanned, I made progress : as I said I saw that the tubes all light up, the front panel illuminates, and :

1) I found where the buzzing sound was coming from : it's not the relay, but rather what looks like an actual/intentional buzzer, an ancient looking electro-mechanical device with a little coil (red in the picture below), that mounted behind the front panel, next to the mains socket, power switch etc.



2) When it buzzes, it also illuminates a big red light on the front panel, next to it, that reads "OVERLOAD" (translated).

Hmmm.... I don't know what is "overloaded" exactly, I didn't find a QR code printed next to it so I could go learn more online about it ?! 

So I assumed maybe a power supply overload ?

So I measured the 6 power rails available at the test port I mentioned earlier.
We have :

+100
+230
+300
-58
-225
-300

So I would power up the instrument, and I would get 3 to 5 seconds max before the buzzer would trigger, forcing me to pull the plug before the fuse blew.
This was just about enough time for the power rails to come up, rising slowly. And just enough time for me to probe the ports with my meter.
So I did this cycled a dozen time. First a round to measure the DC voltages, then another round to measure ripple.

Result : all 6 power rails are alive, and all with low ripple. I mean in the tens of mVrms, so not blatant / catatrophic rectifier or cap failure. It basically "works".

HOWEVER... I noticed that all 6 rails rad exactly 50% of what they are supposed to be !
+300V reads +150V, etc etc.

Hmmm.... that must be some kind of clue.

What could cause this... it must be a cause common to all 6 rails, so quite far upstream.... mains voltage selection ? I set it to 240V not 120V...
But that relay in the area ? Maybe it's partly responsible for the voltage selection, maybe it's faulty, causing the transformer/PSU not to be wired properly ?
I guess that would cause the overload issue once the tubes are warmed up.

Without a schematic at least for the PSU section, it's hard to know.  If anybody has knowledge in thins area, from similar PSU arrangement in some other TE they worked on.. I am all ears. Every suggestion welcome.

In the meantime I guess I will try to trace xwires and reverse engineer the primary side of the PSU. that buzzer, that relay...  there are so many wires going to the transformer, not easy. Will give it a shot anyway.

I have no plans to send any considerable of time nor money on this TE, but I am willing to spend 2 or 3 evenings on it at at least, to see if I can find a simple fix. I really would like to at least be able to stay on without blowing its fuse, so I can at least play with it a little bit.

[Robert763]

Hi Vince,
The relay may be a inrush limiter. However I don't think it is faulty. More likely that the unit powers up under half voltage but somthing breaks down when the in-rush limit closes and full voltage is applied.
I'd pull all the tubes and try again. This may give a clue to the fault location.

Robert.

[Vince]

Thanks Robert, interesting idea this two phase process to bring up the rails.

Yes, let's narrow it down .... so I looked at how this monster is  organized to try to see what's what.
Luckily I find it's very well organized/structured... it's clear.  No confusing spaghetti wiring here.

From what I see, there 4 decks in this thing. Two horizontal and two vertical.



GREEN = Horizontal rear deck is all for the PSU.
RED = Horizontal front deals with the front panel
YELLOW = Vertical left (the TE's left, not the operator's left...) is the attenuator stuff.
BLUE = Vertical right is the trigger/oscillator stuff.

At the edge of the PSU deck, we have 3 big cables with a yellow jacket, secured to the PSU with 3 big ass, exotic, fancy, professional expensive looking connectors...
There is one cable that goes to each of the other 3 decks. It''s limpid, I like that.




I removed the bracket that secures those 3 big connectors, unplugged them... which was surprisingly easy. They are super smootth and come out extremely easily. love them.



So now we have removed power to all 3 decks, the PSU is all alone..... apply power... problem still there.

OK so now we know we can concentrate on the PSU deck alone, making progress ! 

So now I can do as you said, and remove all the tubes in the PSU.

Stay tuned....

[m k]

One must wonder how much Vince's floor space is free.
Or how many items are not stacked.

Maybe one day somebody invents a vacuum cleaner with flat surface.

[Specmaster]

Looks like crap. That screen would be terrible to use....

Yes, the screen is crap, that said it was an early portable scope and as always the early ones of anything, later models always vastly improve on things. I have the same model in my collection, the backlight driver had gone so I sourced an aftermarket driver and patched it in. It works OK but I have better scopes that I can use so I only wanted it in case I ever needed to have a portable scope and meter, then the one tool would suffice. 

[Vince]

MK, no sch thing as free floor space here. My entire life is crammed into the living room for now, so I can barely walk.
Will get better once the garage is built.... in 2 or 3 years all being well. For now I make do, no choice.


Anyway, I just pulled all the tube in the PSU... one hell of a PSU, no less than 15 tubes no less ! 

All with secured with a springy clip, didn't help.
Started by removing the two huge and tall 6080 rectifier tubes, this made some room which helped in getting started removing all the other, smaller tubes.

I just found a new use for ESD foam : stick tubes in it ?
This way I can organize them in the exact same order and layout as they were on the deck, so I can put all the tubes back into place exactly where they were, quickly and reliably...

Anyway. So I fired up the beast with zero tubes and.... no buzzing sound anymore, yeah...

So what next ?

It does not mean any of these tubes are bad... I wish I had a tube tester to rule them out or find a faulty one.
It could just be some passive component that's causing problem, and removing the tubes merely isolated that failed component, keeping it from causing trouble...

So what do I do now ? I put the tubes back one by one ? In what order ? With no schematic to know what each tube does, relative to the others... I am just going blind and potentially doing damage if the sequence is wrong.... I am scared now.
Or, I put them all back in place, then remove them one at a time this time....

I am thinking, further to what Robert said... maybe that relay is used to switch some winding in the transformer, to apply full voltage ? So maybe that winding is kaput, transformer defective, door stop ?
I hope not, but I guess it's a possibility to keep in mind...

[factory]

6080 is not a rectifier, it's probably used as series pass regulator, guess this must have solid state rectifiers somewhere.

Did the 85A2 glow tube light up? might be worth checking the voltage across it and any other glow tube that might be used as the voltage reference.

David

[Vince]

OK, getting somewhere : I tried putting only to tubes back to begin with : the two big rectifier tubes... and the buzzing is back.

OK that's good... I guess that means  the problem lies not in the fancy stuff implemented by the otehr 13 tubes, but rather at the very root of the PSU.. the secondary windings, and, main recttifier tubes, filter caps.

so I will try to look at the wiring of those two tubes more closely.

Maybe they are not in parallel. Maybe there is one for the initial power up at half-voltage, and the second tube sees voltage only once the relay clicks in to switch its dedicated secondary winding.
Something like that.... So maybe that second tube has its own filter cap, maybe that one is bad and pulling too much current.

[Vince]

6080 is not a rectifier, it's probably used as series pass regulator, guess this must have solid state rectifiers somewhere.

David

Oh thanks !

They were so big, and next to the transformer, so I assumed they were rectifying.... that changes my plans then...

Will look for solid state diode, and try to reverse engineer the PSU to some degree, as well as the overload detection system...

[Vince]

Some more : it buzzes even only one 6080 tube. Either of them. If I swap the tubes (still only one in place), buzzes too.

SO that means it overload the PSU regardless of the tube, and with no ohter tube present to drive their grid to actually make them conduct... so that would mean they are shorted.... how can a tube be shorted anode/cathode ? Is that even possible...

I will keep at it...

[factory]

Reading some manuals for the HP 212A* & HP 214A, they both have overload relays to protect the certain tubes, I have two 214As in my collection, the 212A manual states it makes a buzzing sound in overload state, of course the Ferisol might work completely differently. 

*as brutally murdered by oz2cpu on youtube, so his friend could steal the glassware.   :'(

David