Test Equipment Anonymous (TEA) group therapy thread

[GregDunn]

*Hissssssses Python-like in tggzzz's general direction...*

mnem
 

I remember makefiles, where there is a semantic difference between tabs and spaces. You can't see the difference on the screen, but the program does execute very differently.

Python does something very similar: cut and paste a bit of code, and if you don't get the right number of spaces then your code's meaning has changed. It also defeats attempts at cleaning up formatting by using an IDE's autoformat functions.

But I suppose omitting braces/brackets/parentheses is better for one-handed typists.

Python really works best if you have a "smart" IDE to sort those spaces for you; but it's very nice to have an unambiguous visual cue as to the control flow.  It is a problem when transferring the code to another platform.  I always hated that there were multiple ways to indent C/Java code, and none of them made any difference to the control flow despite looking vastly different - if you weren't careful you could completely miscommunicate the structure of the program and still miss syntax errors.  They were less prone to hackers writing indecipherable gibberish than Perl is, but still abused at times.

That said, while Python was a nice thing to have when we were building a complex inventory creation program, it has been infested with object-oriented thinking lately, a similar path taken by C++ or Java - and that's not a good thing (16 years of Java programming taught me that OO is not better than procedural programming, just different).  Though I haven't written any comprehensive programs in it, Go seems to be a step away from OO and back toward C with useful improvements to the base syntax.  You still have to be mindful of formatting it for readability, though no more so than C.

[bd139]

Kicad?

Yep

Jeebus... that's a lot of effort to replace a device you can literally solder-snake inside a bit of heat-shrink tubing on a 99¢ XFMR.    Hats off to you for the courage of your convictions and following through on it, though. 

It is indeed overkill ... This is a "two birds one stone" thing though. It'll fix the alarm and solve another problem I have. At the moment I'm running my radio off a 7Ah SLA pack which I am topping off with a TTi PL154 bench supply occasionally. This isn't ideal as it needs to be futzed around with too often. So the plan is to take the two NEW Yuasa 17Ah batteries I managed to filch the other week and use them in parallel as both a radio and bench lamp supply with this connected permanently for trickle charging and sling them under the bench. Thus the radio and lights will still work when the power is out which happens annoyingly regularly here due to the local substation being overloaded.

I want to make the most of the power outages as the noise floor goes down considerably!!!

[worsthorse]

Yes you're correct, I took another look at the pages and they have been scanned but at a very high DPI, my original posting saying that I doubted it had been scanned I had just deleted from the forum  so now everyone knows I was wrong  , hmm nothing new here, nothing to see, move along now 

Given that the archive.org PDFof the ECG cross reference is almost a gigabyte, I think that "scanned at very high DPI" is a pretty safe guess. 

I have a basic collection of data book scans in PDF rather than in paper format but only because (a) the DTVs of my favorite books are unobtanium or ridiculously expensive (the latter condition essentially creating the former for me) or (b) I know that having two cabinets full of musty smelling databooks from the 80s and 90s would result in a banishment order from SWMBO.

[worsthorse]

Actually KY jelly makes a great lubricator when it comes to trying to squeeze things into tight spots

I was going to offer a reply to this statement but, well, I think it stands on its own.      

[mnementh]

Jeebus... that's a lot of effort to replace a device you can literally solder-snake inside a bit of heat-shrink tubing on a 99¢ XFMR.    Hats off to you for the courage of your convictions and following through on it, though. 

It is indeed overkill ... This is a "two birds one stone" thing though. It'll fix the alarm and solve another problem I have. At the moment I'm running my radio off a 7Ah SLA pack which I am topping off with a TTi PL154 bench supply occasionally. This isn't ideal as it needs to be futzed around with too often. So the plan is to take the two NEW Yuasa 17Ah batteries I managed to filch the other week and use them in parallel as both a radio and bench lamp supply with this connected permanently for trickle charging and sling them under the bench. Thus the radio and lights will still work when the power is out which happens annoyingly regularly here due to the local substation being overloaded.

I want to make the most of the power outages as the noise floor goes down considerably!!!

   If you were stateside, I'd just drop one of these in the post tooya. I have like 3 or 4... A single 1157 bulb in series and you're done.

Plus you get to feel smug in your understanding of how we whooped a battery's ass took tender loving care of our SLAs back in the old days.

mnem

[mnementh]

On my list of Suspect TEA purchases this is 2nd, the First was buying six tubes of KY Jelly at once for a Dopler shift flow meter.

For filling the wells on my Ice Point Calibrator 'allegedly' 

Actually KY jelly makes a great lubricator when it comes to trying to squeeze things into tight spots, eg, sliding long lengths of sleeving over anything, I've used other similar items originally designed for more personal usages, for that before today   

I try NOT to use water-based lubricants on anything that will see more than 5 volts or so; K-Y can stay wet for years inside heat-shrink tubing, and ooze out at the worst possible time.   

Food-grade spray silicone is much better and the carrier evaporates before you're done applying heat. The problem there is that it also encourages heat-shrink tubing to slide off while in service... there's always a gotcha when trying to "make things easier".

mnem
*tzzzzzt*

[tautech]

Indeed. There are very few reputable sellers shifting semiconductors. Usually best to get them from house clearance lots and mixed components lots now and sort and test.

Just spent the last 2 hours designing an SLA battery charger to replace a transformer/diode/capacitor and single resistor in my alarm which packed in yesterday. What a cheap arse hack. Very annoyed.

Replying to myself here. A veritable fuck you overkill charger for my alarm...



This one has CV/CC bulk and trickle charge, crowbar, reverse connection protection, some heatsinking, snubber caps on the bridge rectifier and can be scaled up somewhat for radio operation etc and fits in the cabinet I have already (hence the somewhat larger than necessary size).

Issue one so I expect fire and I put ZERO effort into the silkscreen. £9 for 5 boards from JLCPCB including delivery = win 

Edit: To note I found out when I disconnected it by chance that the existing charger was causing my noise problems on 80m. Assume it was the diodes and the long cable runs! This one should be a little less unfriendly.

LM317, a 10t pot and some protection diodes fits on some Veroboard and smaller than a matchbox.
Let the 317 internal circuitry protect itself.........done !

[bd139]

Need something with a bit more grunt and security than an LM317. This scales down to 100mA charge current and up to 5A and handles high mains voltages, overloads, shorts, reverse connection, source failure and most importantly protects the £1500 worth of kit connected to it if the regulator or battery fails and dumps more than 15V out. Also it handles CC initial charge, dropping to very precise CV mode once it is done. it can withstand a 95A fault current for ~2 seconds (enough to blow the fuse) and a persistent 15A fault current on all battery connected ports.

Incidentally even a nice TI LM317 doesn't last particularly long if you run it at the line eventually going in/out short hence why you tends to see fat 5W zeners across the output on anything worth more than about a quid.

[med6753]

Got the parts to convert the OL-1 tube filament to DCV and the results are "good enough" to consider this project as complete.

As I expected, the DC filament voltage is low under load because I'm rectifying the 6.3VAC filament from the power transformer. I'm getting 5.7VDC which is low but it works...just barely. Initially the waveform was extremely difficult to sync. I changed the sync tube (12AX7) and it's now stable. It's just a matter of selecting tubes which will operate a reduced filament voltage. I could fix this by installing a separate power transformer with regulation but I don't think it's worth the effort. So this guy is done and considering where it started from I can feel good that it's mission accomplished. 

Here's the bridge and capacitor installed just behind the 6X4 rectifier.


And sample waveform at about 5KHz.

[beanflying]

Actually KY jelly makes a great lubricator when it comes to trying to squeeze things into tight spots

I was going to offer a reply to this statement but, well, I think it stands on its own.      

And if it won't stand on it's own take the Blue pill ..... 

I have a few different grades of food grade silicone lube for coffee machine work which works for all sorts of other odd elctro mechanical jobs.

[Specmaster]

@Med, yes you can be justifiably proud of that one, it's come a million miles from where it started, well done sir  , you can give a sigh of relief now the job is done.   

[bd139]

Totally. I am thoroughly surprised at the outcome after seeing the initial picture! Nice job!

[nfmax]

Need something with a bit more grunt and security than an LM317. This scales down to 100mA charge current and up to 5A and handles high mains voltages, overloads, shorts, reverse connection, source failure and most importantly protects the £1500 worth of kit connected to it if the regulator or battery fails and dumps more than 15V out. Also it handles CC initial charge, dropping to very precise CV mode once it is done. it can withstand a 95A fault current for ~2 seconds (enough to blow the fuse) and a persistent 15A fault current on all battery connected ports.

Incidentally even a nice TI LM317 doesn't last particularly long if you run it at the line eventually going in/out short hence why you tends to see fat 5W zeners across the output on anything worth more than about a quid.

Now you need to make sure that your alarm panel won't over-discharge the battery to destruction when (not if) the power stays out for too long, like mine did

I designed & fitted an over-discharge protection circuit which fits between the new battery & the panel.

[bd139]

Dang didn't think of that  ... it does tend to go off if it goes flat though.

issue 2 will have a low voltage cutoff. Thanks for the heads up! I'll get issue 1 working first and thoroughly test it. Should be able to achieve that with a fat low Rds(on) MOSFET, comparator and TL431 off the top of my head.

[Cerebus]

I try NOT to use water-based lubricants on anything that will see more than 5 volts or so; K-Y can stay wet for years inside heat-shrink tubing, and ooze out at the worst possible time.   

Food-grade pray silicone is much better and the carrier evaporates before you're done applying heat. The problem there is that it also encourages heat-shrink tubing to slide off while in service... there's always a gotcha when trying to "make things easier".

Try, if you can find it, Hellerine cable lubricant made by Hellerman-Tyton. I used to use a lot of it back in the days when I used to use Hellerman's neoprene sleeving for wire to board strain relief (audio stuff that was going on the road to be subjected to musicians, DJs, roadies and *shudder* producers). Stays slippy long enough that you can get your work done and when you come back to rework something a week or two later it seems to have all gone. Doesn't interfere with soldering either.

[xrunner]

Totally. I am thoroughly surprised at the outcome after seeing the initial picture! Nice job!

Yes very nice restoration!

[GreyWoolfe]

Actually KY jelly makes a great lubricator when it comes to trying to squeeze things into tight spots

I was going to offer a reply to this statement but, well, I think it stands on its own.      

Astroglide works much better.

[bd139]

Yep I bought some astroglide for a project manager I worked with back in ‘04 as he was a massive wanker 

That’s how you burn bridges 

[mnementh]

Got the parts to convert the OL-1 tube filament to DCV and the results are "good enough" to consider this project as complete...  (SNIP)

   And sample waveform at about 5KHz.

Looking good, buddy. Especially since I can't see that jank BNC connector in this shot... 

mnem
*Waiting for med to come back with a screen-filling pic of JUST that BNC now...* 

[VK5RC]

BNC are fine in my book, a pox on the bugger/s who keeps putting PL259\SO239 on equipment!

[bd139]

I’m in complete agreement there. PL259 can go to hell. BNC or N and nothing else.

[VK5RC]

Maybe SMA for the weird stuff

[worsthorse]

Got the parts to convert the OL-1 tube filament to DCV and the results are "good enough" to consider this project as complete.

And sample waveform at about 5KHz.

That's a fine piece of work.

...and knowing when to say good enough!? Now, that is a skill worth cultivating!

[Neomys Sapiens]

BNC are fine in my book, a pox on the bugger/s who keeps putting PL259\SO239 on equipment!

There is worse. RCA jacks for example. And that nameless one with the flat center contact (rather cheaply made). Here, instruments from that era tend to have 4/13 connectors. Those are 4mm Banana plugs with a 13mm outer contact. Most are well made, even jacks with switches have been seen.  And if none is at hand, one can still use any 4mm for the center contact and get ground from the (hopefully present) separate jack.

[bsfeechannel]

Got the parts to convert the OL-1 tube filament to DCV and the results are "good enough" to consider this project as complete.

As I expected, the DC filament voltage is low under load because I'm rectifying the 6.3VAC filament from the power transformer. I'm getting 5.7VDC which is low but it works...just barely. Initially the waveform was extremely difficult to sync. I changed the sync tube (12AX7) and it's now stable. It's just a matter of selecting tubes which will operate a reduced filament voltage. I could fix this by installing a separate power transformer with regulation but I don't think it's worth the effort. So this guy is done and considering where it started from I can feel good that it's mission accomplished. 

Interesting solution. Guitar amp modders use a trick that is to balance the filament winding using two resistors.



Sometimes they use a potentiometer instead of two resistors to finely adjust the cancellation of hum.