Test Equipment Anonymous (TEA) group therapy thread

[capt bullshot]

The first of the CTI OCXO made its way into one of these DCF77 (http://wunderkis.de/dcf-rcvr/) radio clocks:




This is the smallish PC (an IGEL thin client running headless Ubuntu) that I'm using to log data and remote debug the things.



Some first results:

Captured timestamps of an GPS PPS (green is the OCXO, yellow the originally used TCXO), PPS supplied by a GPS receiver, captured by the radio clocks:



Recorded OCXO / TCXO control voltage (same colours, internal PWM DAC setpoint, not scaled to anything meaningful):



Note the time scale: At night DCF reception is affected by ionosphere reflection, so one gets these larger deviations than while daylight. Due to its stability, the OCXO radio clock is dealing better with that.

Edit: swapped trace colours.

[AVGresponding]

On The Bench This Morning:



AiO CPU Cooler Periodic Maintenance

Y'all may have noticed the dwagon hasn't been around stirring sh...tuff up this morning... I've been a little busy.

Since I have the CPU out of my PC anyways, I figured now was a good time to do periodic maintenance on the cooler in my Gaming/VR machine; this AiO has been in service for ~2 years in various iterations of this build. While these are supposedly a zero-maintenance sealed system, there really isn't any magic here; these coolers suffer all the usual depredations of any liquid-cooled PC.

The only real difference is they design the whole thing as if it were a FRU with a life expectancy of 4-6 years, or just beyond the typical 3-5 year warranty; the system is priced such that it is most cost-effective (in this case, ~US$60 delivered from Amazon) to simply replace it as a unit when any part fails. This mostly means minimizing metal parts that oxidize, and using a lot of the more durable plastics instead. That does not mean it cannot benefit from a little periodic maintenance, however.
 


These gelatinous boogers are the byproduct of the electrolysis between differing metals (copper plate vs aluminum radiator) with a conductive liquid coolant and a little oxidation as it is nearly impossible to purge every bit of air from a system like this. They tend to form in places where the liquid flow changes direction sharply or is stagnant due to some restriction; then they grow to the point they clog things.

There are probably more of these in the end-tanks on the rad; alas, there's little one can do aboot those. Even if the hoses weren't permanently attached with crimped connectors, the small diameter of the hose barbs means there's no real way to get in there and clean anything other than the hose barb itself. The flip side of this issue is that these radiators are usually hugely oversized, and can suffer a lot of lost efficiency before they fail to cool effectively.

You can see here that the rubber divider seal which is supposed to force coolant flow through the micro-fins has softened and ballooned a bit due to prolonged exposure to the coolant; not sure how much that affects efficiency of the cooling.



Removing the impeller housing lets us see the moving bits; the entire impeller is made of rubber magnet material like refrigerator magnets, and the bearing/spindle are ceramic. Interesting that this rubber magnet material is resistant to the coolant. huh.

Nice & clean in here tho, as this is where stuff is always moving.



Now I've got the service-able bits out, I'll take these to the sink and scrub gently under hot water with a natural bristle brush that can get into the micro-grooves. Testing for flow is done the old-school way once all the coolant is scrubbed out under running water; press microgrooves against your lip and see if you can blow air through them. 



Parts have been rinsed with distilled water, and are now ready to go back together. This is done as much by feel as anything; get the rubber divider seal aligned as close as possible by eye, then lay copper plate with divider plate installed against the pump and slide a little until you feel it catch the pre-existing grooves pressed into the surface of the rubber.



Then, we button it up with the original stainless steel screws.

It is important here to thread these just as you would a machine screw; by gently applying pressure and turning the screw backwards until you feel the bump of the thread that's been cut in the plastic by the screw before, then turn forward very gingerly. If you've got it right, you'll feel very little resistance as the screw goes in; if you feel any appreciable resistance, turn it backwards and try again. By developing this skill, you can reassemble screws in almost any plastic without stripping the hole out.

A'aight; time for this widdle dwagon to get back to the build.

Cheers!

mnem

This kind of nonsense is why I built my own.

It's been running for quite a while now, and since I use auto coolant, corrosion isn't an issue. Evaporation is glacially slow, I looked at the header tank the other day and it looks like I may need to top it up in a few more years.

It was also much cheaper.

[kleiner Rainer]

...
 I noticed that the OG3 VR tube for the -150V reference was not ionizing. That's not good. I don't have a replacement. Tried re-seating the VR tube and now good. Time delay kicked in but still have major issues.

med,

I remember a case where a VR tube did not work in a ham radio transceiver (Collins). The result was that the gas in the VR tube was doped with a radioactive material to set the reference voltage. After 50 years, most of that stuff had decayed enough so that the VR tube did not show a constant voltage anymore. Sounds like your problem. Maybe time to build a zener replacement?

Greetings,

Rainer

[Robert763]

...
 I noticed that the OG3 VR tube for the -150V reference was not ionizing. That's not good. I don't have a replacement. Tried re-seating the VR tube and now good. Time delay kicked in but still have major issues.

med,

I remember a case where a VR tube did not work in a ham radio transceiver (Collins). The result was that the gas in the VR tube was doped with a radioactive material to set the reference voltage. After 50 years, most of that stuff had decayed enough so that the VR tube did not show a constant voltage anymore. Sounds like your problem. Maybe time to build a zener replacement?

Greetings,

Rainer

The radoactive component is to provided relaible starting with minimal overshoot rather than set the voltage. This is a particular problem if the tube is in the dark. Two options to fix:
1/Stick a radium luminised watch hand to the outside of the tube (some tubes used external sources a standard)
2/illuminate the tube with a blue or UV LED.

[Robert763]

Heads up for UK ebay members, you might want to watermark your pictures with user name or similar. They have changed the T&C so they have rights to your "content"

"When providing us with content (including causing content to be posted using our Services), you grant us a non-exclusive, worldwide, perpetual (or for the duration of any copyright or other rights in such content), irrevocable, royalty-free, sublicensable (through multiple tiers) right to use the content (including, without limitation, creating and using derivative works). We may in particular use your content, including any photographs you upload, for marketing and promotional purposes."

And
"Further, to the fullest extent permitted under applicable law, you waive your moral rights in the content and promise not to assert such rights or any other intellectual property rights you have in the content against us, our sublicensees or our assignees."

So they can use and sell your photos.

[Cubdriver]

...
 I noticed that the OG3 VR tube for the -150V reference was not ionizing. That's not good. I don't have a replacement. Tried re-seating the VR tube and now good. Time delay kicked in but still have major issues.

med,

I remember a case where a VR tube did not work in a ham radio transceiver (Collins). The result was that the gas in the VR tube was doped with a radioactive material to set the reference voltage. After 50 years, most of that stuff had decayed enough so that the VR tube did not show a constant voltage anymore. Sounds like your problem. Maybe time to build a zener replacement?

Greetings,

Rainer

The radoactive component is to provided relaible starting with minimal overshoot rather than set the voltage. This is a particular problem if the tube is in the dark. Two options to fix:
1/Stick a radium luminised watch hand to the outside of the tube (some tubes used external sources a standard)
2/illuminate the tube with a blue or UV LED.

While doubtless an effective solution, somehow it just seems sacrilegious to put (of all colors) a blue LED inside a 500 series Tek scope.   

-Pat

[Ice-Tea]

Isn’t that a good application for C4? 

Tsk, tsk. No. I expected you to know better.   ANFO/Amatol is the explosive of choice for digging holes.

Good point. I had filed ANFO under "blowing up exchanges"

Five minutes of explosive trenching. It's like watching Bugs Bunny on Amphetamine+Top Fuel+Nitrous.

They should ask them for the next Dune movie 

[tautech]

snip
While doubtless an effective solution, somehow it just seems sacrilegious to put (of all colors) a blue LED inside a 500 series Tek scope.   

-Pat

Maybe but guess what solution Med is likely to choose. 

[Ice-Tea]

OK; got a headscratcher here...

Got another 199C on the bench, this one came with all the bits and bops still in sealed plastic bags so I used some probes I had laying around. Seems to work just fine, and it shouldn't. The BNC shrouds are isolated, doesn't make contact with the probe ground clips. So they might as well not be there. In fact, when disconnecting the clips, the image is still the same. For a "regular" scope, I get it. Ground connection through the mains distribution sockets and all, all tied together. Got it.

But the 199C is 100% floating. How can there be a measurement at all? 

[capt bullshot]

OK; got a headscratcher here...

Got another 199C on the bench, this one came with all the bits and bops still in sealed plastic bags so I used some probes I had laying around. Seems to work just fine, and it shouldn't. The BNC shrouds are isolated, doesn't make contact with the probe ground clips. So they might as well not be there. In fact, when disconnecting the clips, the image is still the same. For a "regular" scope, I get it. Ground connection through the mains distribution sockets and all, all tied together. Got it.

But the 199C is 100% floating. How can there be a measurement at all? 

These isolated BNC should make GND contact with your probes, look inside the plastic connector, there's the metal ring that contacts the slotted cylindric part of the BNC plug. Works for _most_ BNC plugs, but as usual there are ones that do not make proper contact this way.

Are the fluke inputs isolated to each other, or is just the fluke floating? What happens if you disconnect only one or both GND clips? What happens if you add some DC offset to your signal?

[Specmaster]

Heads up for UK ebay members, you might want to watermark your pictures with user name or similar. They have changed the T&C so they have rights to your "content"

"When providing us with content (including causing content to be posted using our Services), you grant us a non-exclusive, worldwide, perpetual (or for the duration of any copyright or other rights in such content), irrevocable, royalty-free, sublicensable (through multiple tiers) right to use the content (including, without limitation, creating and using derivative works). We may in particular use your content, including any photographs you upload, for marketing and promotional purposes."

And
"Further, to the fullest extent permitted under applicable law, you waive your moral rights in the content and promise not to assert such rights or any other intellectual property rights you have in the content against us, our sublicensees or our assignees."

So they can use and sell your photos.

Now thats an interesting topic you raised there, firstly I would fully expect that they will block any images containing any form of watermarking etc, secondly, seeing as almost every item sold on the bay of evil is a commercial product, wouldn't the original makers of those products have something to say about that?

[Ice-Tea]

These isolated BNC should make GND contact with your probes, look inside the plastic connector, there's the metal ring that contacts the slotted cylindric part of the BNC plug. Works for _most_ BNC plugs, but as usual there are ones that do not make proper contact this way.

Maybe, but I disconnected both gnd clips (I think), image was still fine.

Are the fluke inputs isolated to each other, or is just the fluke floating? What happens if you disconnect only one or both GND clips? What happens if you add some DC offset to your signal?

Both channels completely independent from each other. I once had mains voltage inversed on CHA and CHB. Clearest indication they are well isolated or it would have vaporized . DC offset? Haven't tried yet.. Why?

[tggzzz]

Now thats an interesting topic you raised there, firstly I would fully expect that they will block any images containing any form of watermarking etc, secondly, seeing as almost every item sold on the bay of evil is a commercial product, wouldn't the original makers of those products have something to say about that?

If you could reliably detect watermarking, you could reliably remove it.

Some watermarking is invisible, yet survives the image being printed on a inkjet printer and re-scanned into a new image.

Simplest solution would be to add your username near the "interesting" part of an image.

Yes, fleabay is being objectionable, but I'm more relaxed about this than the "f**k off after May unless you give us your bank account".

[capt bullshot]

Both channels completely independent from each other. I once had mains voltage inversed on CHA and CHB. Clearest indication they are well isolated or it would have vaporized . DC offset? Haven't tried yet.. Why?

For AC there might be enough coupling / leakage capacitance to earth within the fluke to show a signal. Any DC offset should vanish then, if it doesn't there's some conductivity between the input GNDs and earth.

[Specmaster]

Now thats an interesting topic you raised there, firstly I would fully expect that they will block any images containing any form of watermarking etc, secondly, seeing as almost every item sold on the bay of evil is a commercial product, wouldn't the original makers of those products have something to say about that?

If you could reliably detect watermarking, you could reliably remove it.

Some watermarking is invisible, yet survives the image being printed on a inkjet printer and re-scanned into a new image.

Simplest solution would be to add your username near the "interesting" part of an image.

Yes, fleabay is being objectionable, but I'm more relaxed about this than the "f**k off after May unless you give us your bank account".

As regards to their payment system, I have to say that so far I'm finding it very good. As soon as I enter details of the shipping and tracking info, they release your money directly into your bank and they deduct their fee at the same time, which at the moment as I'm using the £1 offer, means the full amount less the £1. There are no other fees coming off as there used to be with PayPal then taking their cut.

[Ice-Tea]

Both channels completely independent from each other. I once had mains voltage inversed on CHA and CHB. Clearest indication they are well isolated or it would have vaporized . DC offset? Haven't tried yet.. Why?

For AC there might be enough coupling / leakage capacitance to earth within the fluke to show a signal. Any DC offset should vanish then, if it doesn't there's some conductivity between the input GNDs and earth.

Hmmm, interesting. Will have to check when I can sneak back into the lab. Speaking of Fluke Scopemeters (I think):

[capt bullshot]

Speaking of TEA:



Scene from Doctor Who
https://en.wikipedia.org/wiki/Hide_(Doctor_Who)

[bd139]

Now thats an interesting topic you raised there, firstly I would fully expect that they will block any images containing any form of watermarking etc, secondly, seeing as almost every item sold on the bay of evil is a commercial product, wouldn't the original makers of those products have something to say about that?

If you could reliably detect watermarking, you could reliably remove it.

Some watermarking is invisible, yet survives the image being printed on a inkjet printer and re-scanned into a new image.

Simplest solution would be to add your username near the "interesting" part of an image.

Yes, fleabay is being objectionable, but I'm more relaxed about this than the "f**k off after May unless you give us your bank account".

As regards to their payment system, I have to say that so far I'm finding it very good. As soon as I enter details of the shipping and tracking info, they release your money directly into your bank and they deduct their fee at the same time, which at the moment as I'm using the £1 offer, means the full amount less the £1. There are no other fees coming off as there used to be with PayPal then taking their cut.

Yeah it’s working out fine. Just watch out:

1. Create a new bank account for it.
2. Incoming transactions take 2 working days to clear into your account and are “staged” during that period.
3. eBay will hold an amount from the staged area or any new payments if there is a dispute opened.

Only defence is to remove any high value items from sale if someone opens a dispute of significant value. This forces a stalemate until the dispute is resolved.

[bd139]

Both channels completely independent from each other. I once had mains voltage inversed on CHA and CHB. Clearest indication they are well isolated or it would have vaporized . DC offset? Haven't tried yet.. Why?

For AC there might be enough coupling / leakage capacitance to earth within the fluke to show a signal. Any DC offset should vanish then, if it doesn't there's some conductivity between the input GNDs and earth.

Hmmm, interesting. Will have to check when I can sneak back into the lab. Speaking of Fluke Scopemeters (I think):

Watch The Expanse. There’s a whole bunch of test gear in that 

[ch_scr]

Yesterday my gpsdo module came. No time to waste, put it into an enclosure. Enclosure is sponsored by "5 pound of sh*t in a 3 pound bag". Totally not too small! Totally intended to print bubble out the back to fit buck regulator module from the beginning! Locks, now to wrestle lady heather...
Edit: 3d print files and type of case see posting below

[BU508A]

Yesterday my gpsdo module came. No time to waste, put it into an enclosure. Enclosure is sponsored by "5 pound of sh*t in a 3 pound bag". Totally not too small! Totally intended to print bubble out the back to fit buck regulator module from the beginning! Locks, now to wrestle lady heather...

Your case is looking very similar to this one:

https://www.pollin.de/p/kunststoffgehaeuse-donau-elektronik-kgb21-schwarz-460286

[ch_scr]

Yesterday my gpsdo module came. No time to waste, put it into an enclosure. Enclosure is sponsored by "5 pound of sh*t in a 3 pound bag". Totally not too small! Totally intended to print bubble out the back to fit buck regulator module from the beginning! Locks, now to wrestle lady heather...

Your case is looking very similar to this one:

https://www.pollin.de/p/kunststoffgehaeuse-donau-elektronik-kgb21-schwarz-460286

Yes, but it is this one: https://www.pollin.de/p/kunststoffgehaeuse-donau-elektronik-kgb11-schwarz-460281
If someone wants to share my pain, I have uploaded the modified files.
Still, you will have to heat one "lip" of the holder with a heat gun to lift it up and slip the GPSDO in. After it is mounted to the bottom part of the case. But it does make a small unit 

[Robert763]

Now thats an interesting topic you raised there, firstly I would fully expect that they will block any images containing any form of watermarking etc, secondly, seeing as almost every item sold on the bay of evil is a commercial product, wouldn't the original makers of those products have something to say about that?

If you could reliably detect watermarking, you could reliably remove it.

Some watermarking is invisible, yet survives the image being printed on a inkjet printer and re-scanned into a new image.

Simplest solution would be to add your username near the "interesting" part of an image.

Yes, fleabay is being objectionable, but I'm more relaxed about this than the "f**k off after May unless you give us your bank account".

That was what I menat by watermarking. Sorry if I did not use the right terminology. Ebay currentl allow this as long as it is not contact information.

[capt bullshot]

New page


Edit: screenshot

[tggzzz]

Yes, fleabay is being objectionable, but I'm more relaxed about this than the "f**k off after May unless you give us your bank account".

As regards to their payment system, I have to say that so far I'm finding it very good. As soon as I enter details of the shipping and tracking info, they release your money directly into your bank and they deduct their fee at the same time, which at the moment as I'm using the £1 offer, means the full amount less the £1. There are no other fees coming off as there used to be with PayPal then taking their cut.

I know fleabay can, via one means or another, "deduct" money from my credit card. I can tolerate that. What I haven't assessed is whether they could do something similar with a bank account.

As several people have noted, the workaround is to have a new semi-dormant account solely for that. I haven't opened a bank account in decades, so that's currently an unknown for me. Given other things on my plate ATM, that ain't gonna happen soon