Test Equipment Anonymous (TEA) group therapy thread

[nfmax]

In other news...

installed the Frankenstein-like style built* HP Agilent 8110A 150MHz two channel pulse generator into my rack and ran a quick test with both channels:

I'd like to mod it and make the display switchable on/off to save some of its remaining life -spare parts are available at keysight but over $1k is a bit hefty.   
How could I do that the most easy and less harmful way? Just switch the display modules power? It has a small switch mode ps on board..
Any ideas welcome. 

It has the optional 81106A PLL board in it that improves some performance figures significantly:
*-Thats for a story to be told in another posting...   

Well done! I have one of those, with a very dim but still just about useable display, which has the PLL board, but only one output channel. The front panel moulding is only slightly damaged...

Since this is such a common problem with these units, I suspect someone at Böblingen got the mechanical tolerancieng slightly wrong, leaving too much residual stress in the moulding. They are difficult to get on & off. I fitted mine with a pair of rack handles to provide additional mechanical support to the plastic, by clamping it in place. Rack ears should work too.

Somebody really ought to reverse-engineer the VFD display protocol and make an LCD replacement!

[bd139]

I hate seeing this sort of shit 

[tonyalbus]

You missed the best pro framing hammer Estwing ever made:
https://www.amazon.com/dp/B0000224V9/ref=dp_cerb_1

Agree, that is what a workmans hammer looks like, the others look like they should be in a display case 

Programmers' hammer:

knocking-off time hammer:

mc Hammer

[Wolfgang]

@BU508A Many thanks for your piece of filter material, it arrived today along with your surprise package to keep it company during the posting, it does get kind a lonely I guess in an envelope  I'm sure that I can find a use for them somewhere along the way. Here are a couple of photos with the filter installed and lit, beautiful contrast now 

My pleasure. 

That was exactly my thought. Traveling alone can be sooo boring.      I got an omnibus volume of some electronic parts from Rohde & Schwarz and it contained a huge amount of those µA709 OpAmps and I thought, they could be a nice company. 

Glad, that you've enjoyed it.   

The x-tal tester is looking much nicer with that filter, isn't it?   

... just a question: Does you omnibus also contain vintage transistors (TO-18) ? I am working on some esoteric oscillator project where I need *old* parts that have not been made using modern production processes.

regards
  Wolfgang

Any particular part No.s you are looking for? I have boxes of old TO-18 transistors and probably a few of those Germanium ones like the AC128 (no idea what the case name is).

McBryce.

My project uses old HF and NF transistors like 2N2222A, 2N2857, but also any others that have been made by "old" processes. I found out that these are a lot less
noisy than modern parts when used in high-Q crystal oscillators.

As long as it is pre-90ies all parts are welcome.

regards
   Wolfgang

[bd139]

CDIL make a lot of legacy parts still today on old processes including 2N2222A etc. They're also cheap! Worth having a look there.

[tonyalbus]

Just in Today NanoVNA-H 4 , version 4 up-to 1.5 Ghz and with new firmware 0.7 up-to 2.7 Ghz..
nicely in a proper box..

i need to wait a few hours to play... working right now ;-(

[bd139]

Ooh nice. Let us know how you get on with it 

[xrunner]

Agilent 54622D carbon pad restoration

As I said in a previous post, scrubbing / cleaning the carbon pads again with IPA did not help the conductivity, nor did it even remove much of anything. I say "again" because when I got it several years ago I did the same cleaning. That being said, I tested the keys again after cleaning this time and they did not function any better. They all worked but needed to be pressed "hard".

Taking a look at the pads under the microscope, and comparing them to new pads on a comparison keypad removed from a remote control that was never used, it was clear that a crusty layer needed to be removed. I decided to gently use a 600 grit wet sanding paper. Don't picture Magilla Gorilla with an electric belt sander now!

After the first specks came off (see pic), I cleaned the pad with IPA and it looked much better. I then tried rubbing again with the same light touch, but nothing more to note came off. You would have to scrub a lot harder to get anything more off - the pads are pretty tough - but after the first specks come off there is no need to do any more. And that is good because you are removing the crusty film and nothing more. In other words, these pads are not delicate.

You can check the pics - before and after. Also included is a pic of a good pad from a remote control. The Agilent pad is now has a much smoother surface and presents much more surface area upon a light press of the key.

The carbon pad is ~0.5 mm thick. I tried to measure any difference in thickness before and after the light scrubbing, but I could not detect any measurable difference with my micrometer. And certainly with the naked eye it looks as if nothing is different in the thickness. You could do this many times before you would even measure a difference with anything available to the average hobbyist. The amount of crusty material that comes off is extremely small, and after that the 600 grit wet sanding paper does little more to the pad.

Yet, it makes the pad work nearly like new. The buttons respond as any other "newish" gear I own does, with the slightest push on them. I will look at these buttons in a year (remind me if I forget), and we'll compare the condition of the pads.

[Zucca]

the RigLOL below looks sad and worried now.

[Zucca]

Just in Today NanoVNA-H 4 , version 4 up-to 1.5 Ghz and with new firmware 0.7 up-to 2.7 Ghz..

No tony stop it, I need to keep my money in the bank. How much?  

EDIT: FUUU It's so cheap... see here.

[tonyalbus]

Just in Today NanoVNA-H 4 , version 4 up-to 1.5 Ghz and with new firmware 0.7 up-to 2.7 Ghz..

No tony stop it, I need to keep my money in the bank. How much?  

EDIT: FUUU It's so cheap... see here.

AliExpress  62,- Euro incl shipping , with the plasic casing and the cables, and assasories...  and a vency box
its the 4" version... so for now only up-to 1,5 Ghz.. the FW is defferent from the 2,5" that is open to 2,5 Ghz...
it will maybe also come for this version..

[xrunner]

the RigLOL below looks sad and worried now.

Maybe it should be. 

[Cerebus]

No nail guns in them days and the oldies didn't like the piss thin nails they shot so reckoned they were only good for sissies ! 

Eh? Did you not have real nail guns in NZ back then? Back in those days here in Blighty we had the .22 short powered nail guns and those definitely didn't have the anaemic little brads that pass for nails in modern nail guns. I well remember the first time I saw one in operation, someone putting up studding for a partition wall driving 4" nails straight though 2x4s into masonry in one fell swoop.

[psykok]

I hate seeing this sort of shit 

you can sold fixe that with special aluminium welding rods
 

[Cerebus]

I hate seeing this sort of shit 

you can sold fixe that with special aluminium welding rods

Probably not. That's a die casting and the aluminium alloy used for die casting has a lot of silicon in it. It makes it expand ever so slightly as it freezes - which is great for die casting, but lousy for welding. When welding it as it all freezes the base casting metal expands, the filler metal shrinks and the whole thing either bends out of shape or in extreme cases just cracks again.

[Wolfgang]

CDIL make a lot of legacy parts still today on old processes including 2N2222A etc. They're also cheap! Worth having a look there.

Thanks, I have bought a lot of those, they alle are OK with regards to the datasheets, but the old ones (Motorola, ...)
are still better,. dont ask why. In fact, I have a whole box of 20 different 2222s from the 70ies to now.

[AVGresponding]

This is absolutely normal; a byproduct of oxidation which happens upon exposure to air with ANY baryonic matter to some extent. Carbon is a stable element, but the binders used to keep it on the silicone membrane keypad degrade with age and the surface becomes "glazed".

Add a little electrolysis from the charged PCB contacts and you get this condition which the only real cure is cleaning the carbon pads with alcohol to expose fresh carbon matrix every few years or replacing the silicone membrane itself with a brand-new one. And since they haven't been made in decades, that's pretty hard to make happen; hence the carbon pad repair kits. 

mnem
 

Good luck getting helium to oxidise   :3

Ha-ha... and the list of similar bell-end cases goes on & on. Given the right conditions and enough time etc...

mnem
"The difficult we accomplish as scheduled. Impossible may take slightly longer."

Nope. It takes very special conditions to get helium to bond with anything, and compounds with both helium and oxygen in them are as yet still only theoretical, and rather unlikely to occur in Earth normal air.

https://en.wikipedia.org/wiki/Helium

Yeah, yeah... I'm AWARE. This too is covered under "Given the right conditions and enough time etc..."   As in "theoretical conditions" and "theoretical amounts of time"... say those at either end of the Universe, for example. 

mnem
 

Nope, sorry, I'm going to be pedantic here. It was your statement that "oxidation which happens upon exposure to air with ANY baryonic matter" that I have issue with.
Helium, and for that matter neon, I just picked helium because it's so weird it's one of my favourite elements, does not, and will not, at any point in the future up to and including the heat death of the universe, oxidise.
Oxidation is a specific term and I would argue that unless the process can be reversed (reduction), it's not true oxidation. Cerebus can probably explain it better than me, it's more than 3 decades since I had to make serious use of my chemistry knowledge.
As for "baryonic matter", all normal matter is composed of baryons, in other words, all normal matter in the universe, which does include the likes of neon and helium.

I hate seeing this sort of shit 

you can sold fixe that with special aluminium welding rods

Probably not. That's a die casting and the aluminium alloy used for die casting has a lot of silicon in it. It makes it expand ever so slightly as it freezes - which is great for die casting, but lousy for welding. When welding it as it all freezes the base casting metal expands, the filler metal shrinks and the whole thing either bends out of shape or in extreme cases just cracks again.

You can "stitch" cast ally back together using lead, but on something that small...  idk.
Only seen it used on HGV engine blocks.


EDIT: This link might provide some useful clues/inspiration?  https://www.metalockengineering.com/en/typical-repairs/metal-stitching/

[bd139]

CDIL make a lot of legacy parts still today on old processes including 2N2222A etc. They're also cheap! Worth having a look there.

Thanks, I have bought a lot of those, they alle are OK with regards to the datasheets, but the old ones (Motorola, ...)
are still better,. dont ask why. In fact, I have a whole box of 20 different 2222s from the 70ies to now.

Oh yes indeed I agree entirely there. Even from the aesthetic point they are better

[tonyalbus]

New find ;-) ... i need to stop this..... will arive in 2 days...

Militairy Fluke... waterprove box.

[mansaxel]

knocking-off time hammer:

That must be made in Germany. Every German tool maker I know of have made Flaschenöffner; I've got Rittal and Wera here, and I'm hoping for Wiha next Adventskalender month..

[ArthurDent]

Pffft... amateur. I am a ASE CERTIFIED Munter.    And that's just the ones small enough to fit in a drawer. 

In hammer we trust.

Amet.

mnem

My takeaway from this meme is that Spiderman lacks critical thinking because there is a sink and a shower nearby. 

[Wolfgang]

CDIL make a lot of legacy parts still today on old processes including 2N2222A etc. They're also cheap! Worth having a look there.

Thanks, I have bought a lot of those, they alle are OK with regards to the datasheets, but the old ones (Motorola, ...)
are still better,. dont ask why. In fact, I have a whole box of 20 different 2222s from the 70ies to now.

Oh yes indeed I agree entirely there. Even from the aesthetic point they are better

Aesthetics is one thing, but the major trend that plays a role is chip shrinkage over the years. The 2N2222 was always made by a planar process, but
to save space they later used heavier doping, resulting in more lattice damage, more imperfections and added noise. You can see that by
- ft of newer parts is much higher
- leakage currents of very old parts are (60ies, 70ies) are quite high, then in the 80ies and 90ies they are very low, then rising again
- 1/f noise of new parts is not very good, maybe due to RTP steps

Unfortunately, nobody documents processes anymore, and datasheets get less and less informative.

[mansaxel]

Most WAN stuff these days is a complete bodge I have found even if the protocol stack being used is sensible. Right in the middle of one now which involves 4 parties (DC, us, Amazon, MPLS provider) trying to get a redundant 10GbE link up to two DCs. No one appears to have thrown any failure scenarios at it. "oh so if this port goes down, what happens?". Silence apparently!  I'm not even the bloody network guy but in the kingdom of the blind, the one eyed man is king. And my remaining eye has astigmatism.

Well, I am the network guy ;-)  And I've been around long enough to realise that "MPLS Provider" and "10GE link" are sort of mutually exclusive. It'll be more like a "7,2Gbit link". We found out the hard way. Wavelengths and not only that, waves that occupy an entire ITU Grid channel (so ${CARRIER} can't chaos-TDM them via the Netgear-class switch in the WDM "multiservices" card.. ) are the only way to be certain. That, and nukes from orbit.

As for SDH, I think we should make all system software programmers configure and use a E1 circuit run over some damp ethernet pairs bought from a hamfest, close to numerous squirrels for six months before being allowed near anything production grade.

And all web programmers should be behind 200ms latency, and only be allowed a text console.

[psykok]

Most WAN stuff these days is a complete bodge I have found even if the protocol stack being used is sensible. Right in the middle of one now which involves 4 parties (DC, us, Amazon, MPLS provider) trying to get a redundant 10GbE link up to two DCs. No one appears to have thrown any failure scenarios at it. "oh so if this port goes down, what happens?". Silence apparently!  I'm not even the bloody network guy but in the kingdom of the blind, the one eyed man is king. And my remaining eye has astigmatism.

Well, I am the network guy ;-)  And I've been around long enough to realise that "MPLS Provider" and "10GE link" are sort of mutually exclusive. It'll be more like a "7,2Gbit link". We found out the hard way. Wavelengths and not only that, waves that occupy an entire ITU Grid channel (so ${CARRIER} can't chaos-TDM them via the Netgear-class switch in the WDM "multiservices" card.. ) are the only way to be certain. That, and nukes from orbit.

As for SDH, I think we should make all system software programmers configure and use a E1 circuit run over some damp ethernet pairs bought from a hamfest, close to numerous squirrels for six months before being allowed near anything production grade.

And all web programmers should be behind 200ms latency, and only be allowed a text console.

terminal + vi is everything what  you need 

[Cubdriver]

New find ;-) ... i need to stop this..... will arive in 2 days...

Militairy Fluke... waterprove box.

Damn - that's a nice one!  Any that I've come across looked like they spent their years in a war zone being dragged along behind a tank.  Great score - congrats!

-Pat