TEK TM503 Project

[bd139]

Yes looked it up. That's a bit sad but that's change for you. Added it to the list of things to go and see when I have F-road capability.

Character building is spending two weeks in a tent in Central Asia then two weeks in a van in Iceland with a psychotherapist. Neither a gale or impending armageddon-level event even raise an eyebrow now.

Make sure you check that her future research reports/case notes are appropriately Bowdlerised

I'd rather she published what she was working on verbatim. Because it suggests it's all pointless.

[Specmaster]

@bd139 I bet it was bloody cold in that van at night, it doesn't seem to have any insulation anywhere.

[tggzzz]

@bd139 I bet it was bloody cold in that van at night, it doesn't seem to have any insulation anywhere.

Neither do tents; that's what sleeping bags are for. The van ought to be capable of keeping a howling gale out better than a tent.

I remember camping with my daughter in a howling gale at a cliff top camping site in Morthoe. We chose a windy weekend, since it makes waves for body boarding At dusk another tent arrived on the next pitch, was put up in the wrong direction (across the slope) and not tightened up properly; I pointed out to daughter why that would be an issue. When we woke up next morning, the tent was gone, as if it had never been there. Tee hee; Cassandra rulez.

On another occasion, daughter's mother was once there stuck on the beach while we were body boarding. We came out to hear mother protesting we must be mad; we simply asked "what's the issue, there's no wind in the water, and wetsuits prevent problems above water

[bd139]

@bd139 I bet it was bloody cold in that van at night, it doesn't seem to have any insulation anywhere.

I have a very nice Rab down sleeping bag and merino base layer. I'm good to about -20oC. I was actually running pretty hot!

[MED6753]

Re-cap of the FG501.




The vanquished.



Next step is final calibration.

[vk6zgo]

Neither do tents; that's what sleeping bags are for. The van ought to be capable of keeping a howling gale out better than a tent.

I remember camping with my daughter in a howling gale at a cliff top camping site in Morthoe. We chose a windy weekend, since it makes waves for body boarding At dusk another tent arrived on the next pitch, was put up in the wrong direction (across the slope) and not tightened up properly; I pointed out to daughter why that would be an issue. When we woke up next morning, the tent was gone, as if it had never been there. Tee hee; Cassandra rulez.

On another occasion, daughter's mother was once there stuck on the beach while we were body boarding. We came out to hear mother protesting we must be mad; we simply asked "what's the issue, there's no wind in the water, and wetsuits prevent problems above water

That's what we call a "lazy" wind in Oz.
It doesn't bother going round you, it goes right through you.

[tggzzz]

That's what we call a "lazy" wind in Oz.
It doesn't bother going round you, it goes right through you.

That's a typically blunt and evocative Aussie concept. I like it

[Specmaster]

@bd139 I bet it was bloody cold in that van at night, it doesn't seem to have any insulation anywhere.

Neither do tents; that's what sleeping bags are for. The van ought to be capable of keeping a howling gale out better than a tent.

I remember camping with my daughter in a howling gale at a cliff top camping site in Morthoe. We chose a windy weekend, since it makes waves for body boarding At dusk another tent arrived on the next pitch, was put up in the wrong direction (across the slope) and not tightened up properly; I pointed out to daughter why that would be an issue. When we woke up next morning, the tent was gone, as if it had never been there. Tee hee; Cassandra rulez.

On another occasion, daughter's mother was once there stuck on the beach while we were body boarding. We came out to hear mother protesting we must be mad; we simply asked "what's the issue, there's no wind in the water, and wetsuits prevent problems above water

Yeah, I realise its way better than a tent, but equally Morthoe is way better than Iceland. I used to long distance lorry driving for a while, sleeping in the cab long before proper sleeper cabs were a thing and I used a sleeping bag as well, but I was still bleeding cold in the winter months.

[bd139]

Yeah, I realise its way better than a tent, but equally Morthoe is way better than Iceland. I used to long distance lorry driving for a while, sleeping in the cab long before proper sleeper cabs were a thing and I used a sleeping bag as well, but I was still bleeding cold in the winter months.

Sleeping bags have somewhat improved since then.

Mine demonstrated by the eldest who was doing an impression of a sandworm from Dune at the time

Screenshot 2024-07-03 at 14.01.43.jpeg

[tggzzz]

Mine demonstrated by the eldest who was doing an impression of a sandworm from Dune at the time

I've been like that in a ridge tent in Iceland

[MED6753]

The calibration of the FG501 went smoothly with just minimal adjustments required. This unit is now done.



Some sample waveforms. The output impedance of this function generator is 50 ohms so the 2465 vertical was set to 50 ohms.





Next up is the PS503 Power Supply. I will be doing some extensive checking with an ohm meter first and possibly re-cap it before applying power.

[MED6753]

Re-cap and initial check out of the PS503 Power Supply.

Re-capped.



After replacing the electrolytics and before replacing the tants decided to do an initial test. The supply came up fine, for about 5 minutes, then the positive supply went into current limit. C155, a 100uf/20v tant, got warm but did not flame because there wasn't enough available current. Verified shorted. Bloody tants. This is why I always replace them.



Replaced caps. One of those blue tants was the shorted culprit.



The supply works. This is in dual track mode. The fixed 5V supply also works.




Left to do is a calibration. I discovered while replacing the capacitors that the positive supply had a significant repair sometime in the past. Several transistors and diodes were replaced and there was a trace repair on the board.

[MED6753]

Finishing up this project.

I made an executive decision to not replace the capacitor shown, plus one under it, on the TM503 supply at this time. In order to get the old ones out it will take major disassembly and lots of cursing to get the tabs unsoldered from the board. I have replacements in stock just in case.



The calibration of the PS503 was quite simple with no issues. So this project is complete. At some point I'll find another module for the vacant slot.

[Zenith]

I made an executive decision to not replace the capacitor shown, plus one under it, on the TM503 supply at this time. In order to get the old ones out it will take major disassembly and lots of cursing to get the tabs unsoldered from the board. I have replacements in stock just in case.

Probably a wise choice. I know from very recent experience that desoldering those cans can be a PITA. I used a 30W iron and a Soldapult and also a desoldering station. I found the desoldering station was the best, but it took ages and there's the risk of damaging something.

There might be a deluxe way of doing it, using a custom hot air head or something.

[mnementh]

Finishing up this project.

I made an executive decision to not replace the capacitor shown, plus one under it, on the TM503 supply at this time. In order to get the old ones out it will take major disassembly and lots of cursing to get the tabs unsoldered from the board. I have replacements in stock just in case.

(SNIP)

Not gonna take a Sawzall to it like that other one, eh...?

mnem

[MED6753]

Finishing up this project.

I made an executive decision to not replace the capacitor shown, plus one under it, on the TM503 supply at this time. In order to get the old ones out it will take major disassembly and lots of cursing to get the tabs unsoldered from the board. I have replacements in stock just in case.

(SNIP)

Not gonna take a Sawzall to it like that other one, eh...?

mnem

If sometime in the future I have to pull it apart and replace the sawzall might be in play again.

[mnementh]

mnem
this is where I usually put something pithy.

[vk6zgo]

mnem
this is where I usually put something pithy.

You are always "taking the pith"!

[ch_scr]

I present to you: sanitized DM501 DC input "core circuit" schematics. AC circuits omitted for clarity.
Try reading the explanations of DC / Ohms / Integrator (beware of typos) again with this in hand...

AnalogInput_TekDM501.gif

The various bias circuits and current mirrors (and care taken upon them - or not) on display are surely a sight to behold!

[MED6753]

Decided to revisit the DM501 DMM because I have a hunch as to what might be the cause of inaccurate DCV readings. But first....what is underneath that smaller raised board?



A freaking sea of tants. About 9 of them. Did a quick check for shorts but found none. But if I successfully repair this unit those tants will be replaced.



My hunch centers around the input divider shown on the schematic as R10.



It is a sealed unit with Tek P/N 307-1015-00.



It even has it's own TekWiki page which is very interesting.

https://w140.com/tekwiki/wiki/307-1015-00

In order to check it accurately I decided to pull it and check the internal resistors. Mine was good except for the 9MEG. Measured 8MEG. Clearly out of spec for a 0.1% resistor. I verified this on several DMM's to make sure I wasn't being fooled.



I don't have a 9MEG/0.1% resistor so I'll have to order one. Then I'll have to cut open the divider assembly and complete the repair. Will it fix the DCV issue? I'm hopeful but we'll see.

[AVGresponding]

That's a strangely precise and accurate deviation. Are you sure it's not meant to be an 8M with another, extrnal 1M in series, as a early/late variation?

[ch_scr]

I would suggest a "divide and conquer" strategy and measure the (attenuated) input voltage before the integrator at point "circle 1" / before R140 to help chase it down. But I can see how that would be quite annoying without making up something like an extender card first, or at least soldering on a twisted pair of thin wires to feed outside. It's strange for sure for the resistor to be this close to bang-on 8 Meg, it would have be the "coincidence of the century" for it to have drifted there.
You should be able to measure the DM501 supposed 10Meg input impedance from the outside with another DMM; putting the DM501 into DCV mode the other one to resistance (manually switch the ranges as well); that should be an easy test to prove/disprove Spara's theory.

[MED6753]

That's a strangely precise and accurate deviation. Are you sure it's not meant to be an 8M with another, extrnal 1M in series, as a early/late variation?

Agree with the exactly 1MEG deviation. If it is indeed an 8MEG resistor I would expect it to carry a different P/N than the 307-1015-00 which clearly has a 9MEG resistor. I ordered a 9MEG/0.1% and we'll find out when I open the assembly.

[tggzzz]

My hunch centers around the input divider shown on the schematic as R10.

It is a sealed unit with Tek P/N 307-1015-00.

That indicates it is a network, so the tracking (50ppm?) is probably as important as the 0.1%.

[ch_scr]

My hunch centers around the input divider shown on the schematic as R10.

It is a sealed unit with Tek P/N 307-1015-00.

That indicates it is a network, so the tracking (50ppm?) is probably as important as the 0.1%.

If you look at the actual pictures though, it sure looks like they "just" put a bunch of 50ppm/0.1% resistors in a box.
To which degree they tested (tempco, aging/stability) these before / after assembly is up for speculation.
I'd say make them, then test the tempco and throw away the "bad tracking" outliers;
vs. individually tempco measuring the resistors and then matching them up into individual "good tracking" networks.
Depends on how good you need it; how evenly/randomly the resistors behave; what they cost vs. the effort involved, ...
Since they give a (put simplified) 25°C +-10°C window to be in-cal to 0.1%, that's like +-0.01%/°C so 100ppm/°C (for the whole device).
On might question if they did any tempco test on the networks at all, then?