Speaking of cutting your finger off............. One of my trade school instructors told me this story about when he first started out in electronics in the forties, and had his own radio repair shop. A customer brought him a dead radio, and said he'd be back to pick it up in a week when he got paid. After a few days my teacher began to smell something really foul in the shop. He followed his nose to a particular radio, removed it from the shelf, put it on the bench, removed the back of the set, and the funk hit him. Vacuum tubes in those days had grid caps on top of the tube, as the grid is high impedance, and the glass surface of the tube would give lower leakage, than running the grid connection through the bakelite base. The grid return resistor on one of the IF amps had opened up, allowing the tube to saturate and killed the radio. The set owner was poking around in the set, and noticed that if he touched the grid cap and the chassis, his fingers formed a makeshift grid return resistor, and made the radio work again. So this customer went to the kitchen, opened the fridge, cut a strip of liver, draped it around the IF amp tube across the grid cap, touching the chassis, and secured it with a piece of string. The radio played for a few days until the liver dried out. At this point the customer decided to take the radio in for repairs. So yeah cutting off your finger will work, but the garage will be stinky in a few days. It might even mask the smell of the selenium rectumfrier that you tried to smoke!
I grew up in the 1960s and as soon as Dad got done with his Popular Science (or before he got home from work) I would read them from cover to cover. I always liked the "Gus Wilson's Model Garage" stories. I actually learned a great deal of troubleshooting from those - just didn't realize how much until a few years later when I was an electronics tech in the Air Force. This kind of problem was right out of those stories. Well done.
Wow, this brings back memories! My first job in electronics was with a small contract manufacturer in Portland, OR, and we built FG502s, LA501s, and SG502s, if memory serves. Tek provided the parts, we provided the labor. I suspect you may have a prototype, an early production model, or maybe even a "gray" unit, built on the sly by someone at Tek. I say this, because I seem to recall that the units we built had gold-plated traces, especially at the contacts in the cam switches. Not sure if the boards had solder mask, either. Good job on the repair! You were certainly right to suspect the IC sockets. Those were kind of crap, and the later machined-contact sockets were far more dependable. And the op-amps were 741s! That shows just how old this design really is. Thanks for the stroll down memory lane! Keep up the good work!
Those TM-5xx modules are great, I spent a few months buying and restoring about 20 different items, then I gave them all with some mainframes to a friend for his lab. I prefer smaller individualized TE, but the modules are great if you have a permanent lab. As you've been finding, the faults are mostly (1) mechanical, (2) tantalums, and (3) oxidized contacts. *Very* rarely did I ever have to replace a semiconductor, usually it was a 1N914, 1N4001, or similar discrete diode - but some of the older 70s transistors were less than reliable. Fortunately, these have very few Tek-specific semiconductors, unlike the scopes of the day. Having a 3D printer to re-print some of the brittle plastics is a major advantage, especially on the units that have rear-side PCB alignment clips, which always fall apart. I recommend printing in nylon if you have the capability, otherwise 'durable' resin prints work great for knobs and pull-handles. Love the channel and the content - Cheers!
I picked up a couple modules when the lab down the hallway was being decommissioned. They worked at first but then started failing a couple weeks later. Thanks for the tips!
Awesome! I love troubleshooting faults like this. For some reason 9/10 it ends up being something stupid like this. I've seen Tektronix make similar poor layout decisions that can lead to issues like this. And I can't believe they didn't tin the wires for the leads to that pot. I've been buying a lot of this kind equipment lately and usually in a non-working state. For me, repairing it is as much fun as using it!
The two outside solder connections of the frequency pot looked a little dodgy as well...the one on the left looked like the solder had separated from part of the wire and the one on the right looked like not all of the wire strands had made it into the PCB hole. Or it could have just been an optical illusion! ;-) Always nice to bring an old dog back to life! 73 - Dino KLØS
Q515, Q506. The collectors are tied to the case. Between the collectors are a resistor and three diodes to set the AB bias of the next stage. Shorting the collectors together shorts out the resistor and diodes and converts the output to a class B amplifier. I was thinking it was the output amplifier, but you started probing the input signal and it was also distorted. Even I went and started to blame the oscillator circuit.
Those push on ones normally are made from a brass alloy though, at least for almost all the TO39 can ones I saw that were tubes. The aluminium ones generally were extruded and anodised, generally with a red dye as coat.
The distortion at the input of the amplifier was smaller relative to the signal than the distortion at the output. So I assume the distortion was produced in the amplifier and was lead to the input by the feedback (R566). Connecting collectors (heat sinks) in this case Q515 and Q506 would produce crossover distortion.
It looks like the solder only adhered to the outer strands of the wire in combination with being cold. When the wires were disturbed the outer strands took all the stress of the tugging and bending when opening the case(?) When they eventually broke, the inner ones don't have anything to hold onto, there was never solder soaked in the middle. In another words, the strain was not relieved up into the cable, it was concentrated at the board level, leading to premature breakage. It's always good to flow sufficient solder into the middle of the wire. If the wire has poorly plated strands, pre-fluxing after stripping may be necessary, which they obviously didn't do.
Usually heat sinks are anodized making their surface pretty much non conductive but I see somebody already mentioned that. I've always liked Tektronix equipment but I have always wondered how they got those dinky switches to work and work over time.
Wet slug tantalum, with the glass seal, will run till the case rots out, generally from the outside. The ones with the rubber seal will all fail after 20 years.
Throughout half of the video, I was shouting in the voice that it's a problem of bias in the preamplifier cascade transistor Q506-Q515. In other words, the circuit CR520-CR522-CR524 was short-circuited.
Heat shrink tubing is thermal insulation and will defeat the heatsink. It would be far better to drop a spot of RTV or hot glue between them. Evidently, the short circuit was loading the prior stage.
Oh yes, the old magic-finger trick… always joked that I should cut of my finger and glue it to that problematic spot. Not always easy to find the fault that your finger fixed.
I'm surprised at the poor quality of the soldering in this Tektronix instrument. It must have been assembled on a Friday afternoon in the days just before automated assembly of PC boards. As every good robot knows: "You can't trust these humans. They are always thinking about their next beer. If you want something done right you have to do it yourself :-]
Nah, that's not Tektronix... look at the other joints. I've taken apart lotsa these plugins and the factory solder work from vintage Tek is always second to none. Someone else has had their way with this poor creature. That's never a good sign with used electronics coz you know "If I can't fix it I can fix it so it can't be fixed" which is how it ends up in your hands. And speaking of hands Mr Hands, when do we see the rest of you???
Speaking of cutting your finger off............. One of my trade school instructors told me this story about when he first started out in electronics in the forties, and had his own radio repair shop. A customer brought him a dead radio, and said he'd be back to pick it up in a week when he got paid. After a few days my teacher began to smell something really foul in the shop. He followed his nose to a particular radio, removed it from the shelf, put it on the bench, removed the back of the set, and the funk hit him. Vacuum tubes in those days had grid caps on top of the tube, as the grid is high impedance, and the glass surface of the tube would give lower leakage, than running the grid connection through the bakelite base. The grid return resistor on one of the IF amps had opened up, allowing the tube to saturate and killed the radio. The set owner was poking around in the set, and noticed that if he touched the grid cap and the chassis, his fingers formed a makeshift grid return resistor, and made the radio work again. So this customer went to the kitchen, opened the fridge, cut a strip of liver, draped it around the IF amp tube across the grid cap, touching the chassis, and secured it with a piece of string. The radio played for a few days until the liver dried out. At this point the customer decided to take the radio in for repairs. So yeah cutting off your finger will work, but the garage will be stinky in a few days. It might even mask the smell of the selenium rectumfrier that you tried to smoke!
that's a great story, very gross 😅
Brilliant story!😂
Rectumfrier😂😂, haven't heard that since my earliest days in the tv trade, great stuff!.
I've never heard of bigger idiots, that's really remarkable! What part of the country do they live in exactly?
I grew up in the 1960s and as soon as Dad got done with his Popular Science (or before he got home from work) I would read them from cover to cover. I always liked the "Gus Wilson's Model Garage" stories. I actually learned a great deal of troubleshooting from those - just didn't realize how much until a few years later when I was an electronics tech in the Air Force. This kind of problem was right out of those stories. Well done.
My uncle had a auto shop and my dad would call him Gus
Wow, this brings back memories! My first job in electronics was with a small contract manufacturer in Portland, OR, and we built FG502s, LA501s, and SG502s, if memory serves. Tek provided the parts, we provided the labor.
I suspect you may have a prototype, an early production model, or maybe even a "gray" unit, built on the sly by someone at Tek. I say this, because I seem to recall that the units we built had gold-plated traces, especially at the contacts in the cam switches. Not sure if the boards had solder mask, either.
Good job on the repair! You were certainly right to suspect the IC sockets. Those were kind of crap, and the later machined-contact sockets were far more dependable. And the op-amps were 741s! That shows just how old this design really is.
Thanks for the stroll down memory lane! Keep up the good work!
plenty of TEK stuff with tin-lead boards
As soon as you said tapping around the knob I knew what your problem was…mine had the exact same issue!
Those TM-5xx modules are great, I spent a few months buying and restoring about 20 different items, then I gave them all with some mainframes to a friend for his lab. I prefer smaller individualized TE, but the modules are great if you have a permanent lab. As you've been finding, the faults are mostly (1) mechanical, (2) tantalums, and (3) oxidized contacts. *Very* rarely did I ever have to replace a semiconductor, usually it was a 1N914, 1N4001, or similar discrete diode - but some of the older 70s transistors were less than reliable. Fortunately, these have very few Tek-specific semiconductors, unlike the scopes of the day. Having a 3D printer to re-print some of the brittle plastics is a major advantage, especially on the units that have rear-side PCB alignment clips, which always fall apart. I recommend printing in nylon if you have the capability, otherwise 'durable' resin prints work great for knobs and pull-handles. Love the channel and the content - Cheers!
I picked up a couple modules when the lab down the hallway was being decommissioned. They worked at first but then started failing a couple weeks later. Thanks for the tips!
What a cool find and fix! Things always seem so simple when they aren't a mystery any more. My money was on the IC socket being dirty.
Awesome! I love troubleshooting faults like this. For some reason 9/10 it ends up being something stupid like this. I've seen Tektronix make similar poor layout decisions that can lead to issues like this. And I can't believe they didn't tin the wires for the leads to that pot. I've been buying a lot of this kind equipment lately and usually in a non-working state. For me, repairing it is as much fun as using it!
The two outside solder connections of the frequency pot looked a little dodgy as well...the one on the left looked like the solder had separated from part of the wire and the one on the right looked like not all of the wire strands had made it into the PCB hole. Or it could have just been an optical illusion! ;-) Always nice to bring an old dog back to life! 73 - Dino KLØS
Q515, Q506. The collectors are tied to the case. Between the collectors are a resistor and three diodes to set the AB bias of the next stage. Shorting the collectors together shorts out the resistor and diodes and converts the output to a class B amplifier. I was thinking it was the output amplifier, but you started probing the input signal and it was also distorted. Even I went and started to blame the oscillator circuit.
The aluminum heat sinks were anodized so that would provide some insulation, but over time friction could wear through that oxide layer.
Those push on ones normally are made from a brass alloy though, at least for almost all the TO39 can ones I saw that were tubes. The aluminium ones generally were extruded and anodised, generally with a red dye as coat.
The distortion at the input of the amplifier was smaller relative to the signal than the distortion at the output. So I assume the distortion was produced in the amplifier and was lead to the input by the feedback (R566).
Connecting collectors (heat sinks) in this case Q515 and Q506 would produce crossover distortion.
It looks like the solder only adhered to the outer strands of the wire in combination with being cold. When the wires were disturbed the outer strands took all the stress of the tugging and bending when opening the case(?) When they eventually broke, the inner ones don't have anything to hold onto, there was never solder soaked in the middle. In another words, the strain was not relieved up into the cable, it was concentrated at the board level, leading to premature breakage. It's always good to flow sufficient solder into the middle of the wire. If the wire has poorly plated strands, pre-fluxing after stripping may be necessary, which they obviously didn't do.
so the problem was propagating backwards up the signal chain? Wacky!
can you re-introduce the problem by shorting the heatsinks again?
Yes
@@IMSAIGuy oh my.
Happens to the best! I *always* tin the wires before soldering.
Usually heat sinks are anodized making their surface pretty much non conductive but I see somebody already mentioned that. I've always liked Tektronix equipment but I have always wondered how they got those dinky switches to work and work over time.
Would have liked to have seen you jumper those guys while watching the scope.
Yes, tin your wires and don't stop there!
Wet slug tantalum, with the glass seal, will run till the case rots out, generally from the outside. The ones with the rubber seal will all fail after 20 years.
Throughout half of the video, I was shouting in the voice that it's a problem of bias in the preamplifier cascade transistor Q506-Q515. In other words, the circuit CR520-CR522-CR524 was short-circuited.
you need to shout louder next time so I can hear you
@@IMSAIGuy 😁
Great fix...cheers.
Where do you get the schematic for these tests modules?
w140.com/tekwiki/wiki/Main_Page#tab=Modular_Instruments__28TM500_29
@@IMSAIGuy great thanks
Heat shrink tubing is thermal insulation and will defeat the heatsink. It would be far better to drop a spot of RTV or hot glue between them. Evidently, the short circuit was loading the prior stage.
Nice fix ;)
Oh yes, the old magic-finger trick… always joked that I should cut of my finger and glue it to that problematic spot. Not always easy to find the fault that your finger fixed.
Nice
Pin 1 of that RCA 745 chip looks highly dodgy.
it does, doesn't it
I'm surprised at the poor quality of the soldering in this Tektronix instrument. It must have been assembled on a Friday afternoon in the days just before automated assembly of PC boards. As every good robot knows: "You can't trust these humans. They are always thinking about their next beer. If you want something done right you have to do it yourself :-]
crashing bias
Nah, that's not Tektronix... look at the other joints. I've taken apart lotsa these plugins and the factory solder work from vintage Tek is always second to none. Someone else has had their way with this poor creature. That's never a good sign with used electronics coz you know "If I can't fix it I can fix it so it can't be fixed" which is how it ends up in your hands. And speaking of hands Mr Hands, when do we see the rest of you???
It looked factory to me. the wire was really hard to tin, so maybe they just got a bad batch.