Test Equipment Anonymous (TEA) group therapy thread

[med6753]

Could those insulators for the vertical transistors in Med's troublesome 547 also be the same stuff, or not? 

David

I'm not sure if they are BeO washers or not. The replacement transistors came with the insulators already installed so I just simply swapped them over.

[Cerebus]

I did some (careful!) bending of those contacts and putted it on a pegboard without copper overlay (in Germany we are calling them "Knäckebrot"  ).

Come breakfast time, when you're bleary-eyed and still half asleep, please remember not to drape it with ham and cheese and bite into it. 

[Cerebus]

On a more positive note I am at 49 lbs weight loss.

22 kg? Bloody hell, well done.

[Cerebus]

Just checked on my JLCPCB order - checking on progress of things I've ordered is my version of doomscrolling - and as predicted the HASL board has been made, packed and is waiting for carrier pickup. Pretty good - ordered about 2PM yesterday and ready to be shipped a bit over 24 hours later. Bloody excellent when you consider that I paid £1.59 plus postage for the privilege of having 5 boards custom made for me, less than I'd pay for 2L of milk or a loaf of bread. I still can't get over their pricing for basic boards.

The ENiG board has only got as far as solder mask application but even that will probably be made and on its way by the time I get up tomorrow.

[mnementh]

   https://a360.co/37CE8El

Today's Project: Cheap RF Active Probe

So a while back, when I was tinkering on that Pioneer with the dead CPU/16MHz oscillator, bd mentioned the Bob Pease FET probe. I had built such a probe forever ago in school; I also remembered how sensitive it was to assembly technique and resultant stray capacitance.



A little online research revealed this article from ELEKTOR; I actually recognized the diagram from a discussion I read some while back on another random trawl. More research found a number of fleaBay listings for probes based on the design, and for the ~$8 difference between it and just getting the parts I didn't have on hand, I let laziness prevail and picked the one with the most positive feedback which was also one of the least expensive.

While that was winging its way to me, I found an eevBlog thread on the device, which pretty much concurred with my thumbnail assessment; it is far from perfect, but definitely a usable tool as long as you grok that it's really best up to ~500MHz, or 1/3 the advertised max frequency.

Yes, Wolfgang, I read your article as well; I've been poking around your site for years. I just wasn't interested in the assache of building something I could buy 90% ready to use for 20 bux delivered.



Here's what arrived yesterday; the probe and a 300mm soldered-shield RG174 SMA cable.    I liked the compact design, and already had a battle plan for deployment. I did not like the crazy long pogo-pin probe, but I had plans for that as well.

   

This was the plan; move the pogo-pin onto the GND pad, backed away and shortened up to minimize common-mode noise, while the active pad on the input gimmick capacitor gets a nice short little bit of nickel-plated sewing needle for a probe.  Testing with my bench supply showed it draws ~20mA; this should be within the capacity of an alkaline A27 battery at 60-80mAH. Shopping on Amazon found a 5-pack of cheap Chinese A27s for 5 bux; next best with next-day was a 4-pack of most likely counterfeit Energizers (bulk packaging, no expiry date) for $10. I went cheap. 



"No battle plan survives first contact with the enemy..." is especially so when the enemy is yourself; while doing the above testing, that pogo pin snagged on the dwagon's tough hide and broke.  So I used the damaged pin on the center, as it is finer and sharper than my bit of needle. I have some slightly more robust pogo pins on the way for a GND probe; we'll see if they arrive Friday as promised. While I had the thing unsoldered to fix the damaged pin, I decided to go ahead and grind a more probe-ily profile to that PCB and started designing the shell/battery holder based on dimensional drawings for the A27 battery. This was Rev0.5; basic design is there, I just needed to make allowances for component clearance.



This was roughly the point where it all fell apart; by the time the batteries got here I already had the holder made.  I figured that since I have no intention of ever putting this away with a battery in it, a switch was just an invitation to be stoopit and do so anyways; so decided to apply KISS principle and make popping the battery in/out super-easy. I thought I was quite clever when it popped right in and held in place with the (+) peg in a cup exactly as intended.

Then I did load testing and found it sucked the A27 battery down in a matter of minutes... very annoying. Sometimes, all the info you can find isn't enough; my estimated battery life wasn't even close. So... back to the drawing board and a redesigned shell just a hair bigger for an A23 cell.



While that printed, I poked around in my "harvested SMDs" drawer and found this delicious emerald green LED; a little trial/error with a pot showed that a 100K ballast would make a very usable gas gauge with considerable difference between ~7V (the VReg's cutout voltage) and 12.5V as measured on a brammy new battery. Maximum current draw was only 100uA at 12.5V applied, right down to ~40uA at 6V, which was where output signal turned to noise. Adding it was dead-easy with a 0603 resistor and some creative soldering. Here you can also see how I leveraged the mass of the SMA connector to serve as anchor for the batt (-) spring.



Here I've printed up the last parts and made a battery terminal to match; you can see the little dimple I designed in as a window for the LED.



Next on the agenda will be to do some testing, once I find appropriate adapters. Maybe I'll faff around with it on my TinySA/nanoVNA as well. 

And maybe I should add a probe cover; that little fucker is poky! 

   

Added! And updated the Fusion Link/Pic!

mnem
 

[Specmaster]

Later than I'd hoped, but here at last are the photos of the TTi SM630, it is just as if it had just been made at the factory, inside and out. Needs a little cleaning on the case, but thats to be expected. Going by the date code on the chips, and there are many of them, this was made in 1994 or later. The backup battery looks new and judging by the soldering on the back of the board, those three connections are all new and shiny, unlike the others which have that aged dull appearance.

I have run the scope through its paces using the Heathkit IG4505 calibrator, and it hits the right frequency and periods as per calibrator, so no wuckers there. Multimeter part also meets the requirements, agreeing with both the AD584-M but also the current ratings against the Mastech MS7221 calibrator and the frequency counter is pretty much on the money too.

The backlight, as already mentioned in a previous post, is useless. It is one of those reflective types with a white panel and maybe unobtainuim, will speak to TTi about this aspect. If it is as I suspect, unrepairable, then it really becomes a SM620 which never had a backlight anyway.

I'll have to do the function generator in another post.

[Vince]

GOOD TEA day for me !!!! 

A friend surprised me and paid a quick visit to give me some electronic stuff he salvaged from work. He works for "Orange" , ex "France Telecom", the main phone operator in Frog land. He is out all day in the field maintaining the phone and Internet infrastructrure. Every time he tosses some defective and/or obsolete piece of gear, he brings it to me instead of tossing it in the trash as they normally do ! 

So today he brought a phone line test set, very orange as you can see, to test analog lines. Very rugged and very much water proof. Quick tear down pics.
Also a couple very big boards, don't know what they do.... very heavy, measured the PCB thickness with the calipers, at 2.5mm ! 
Both boards have roughly the same topology. 2/3 of the space is occupied by what looks like  a bunch of I/O stuff. Then last third is the main CPU coordinating all that. Lots of big BGA chips, ALCATEL, BROADCOM mostly.
Power supply is a third party daughter board as you can see. On one of the boards the PSU released the magic smoke it seems !

Then the pièce de résistance.... a new scope !!! 

But something really special ! Not just yet another 2000 series Tek, or yet another hollow state 500 series Tek... no... still a Tek, but something very special !

A 11000 series scope !    
I know, sub sampling scope, useless blah blah blah... don't care, always wanted one in my "collection" ! They are so different from all the other scopes, so weird in their UI, so advanced for their time... so cool looking... looked like a million dollar scope. Also very rare on the used market today ! Last one I saw in the local / French ad site was 18 months ago ! Contacted the seller within an hour but of course it was already sold...

So, 18 months later, another one shows up ! Was complete with all 3 plugins, in good cosmetic nick, and seller even pretended it was working just fine on all channels (even though "of course", as always, the pic he showed was of the scope powered off...). Said he used it for 4 years and developed a few boards with it, said "it's big and old but works fine so great for beginners with not much money to put in their first scope".....For sale at 90 Euros ! For a working one complete and in good nick ?!   
Lucky for me, it was for local pick-up only.... so would deter many. I therefore had much less "competition"... obviously the guy was hardly local to me, half way across the country, in Paris !      But he was in the west of Paris and I knew several people in that area, from the Renault Safrane forum I frequent. One of them went to pick the scope up... when the seller wanted to demonstrate to him that it works.... bloody scope did NOT power up, even though he checked it just a few days before, typical ! 

Seller was apparently very embarrassed and said to my friend TAKE IT, it's free, sorry for the trouble, I apologize !     

COOL, a free 11K in perfect nick, and some troubleshooting fun ahead to get to know it intimately, what more could I ask for ?! 

But still, it was in Paris... not going to drive there, way too long of a trip, my right leg hurts real bad after only 45 minutes of driving, having to "regulate" the gas pedal / speed. Need cruise control... hear it can be retrofitted, my model year is half wired for that. It's not too much of a job to retrofit apparently.

Anyway.... then another member also near Paris, from that same forum had planned to come in my area for his summer vacation, and offered to bring the scope to my door step for free. So, 6 weeks after buying the scope, today it got delivered ! 

First attempt : scope powers up, fan spins, LEDs on the front panel light up and dance as you would expect but.... nothing on the CRT display ! Not a sausage...

Turned it off, left it alone for a few hours.. powered it up again... and this time it starts up just fine, I do have a display this time, I can see the self test being performed, and it passes with flying colours, not a single error ! Played with the thing for a few hours and it all just seem to work just fine, as the seller said !!! 

Display might be very slightly dim, if I am being picky, but other than that...

I do notice though that the probe calibrator output does not appear to work. I get no signal out of it, just a +300mV DC level. So that's great, some trouble shooting for me, on something simple that's totally harmless ! Of course I assume there are no schematics for this thing.... though maybe there is, seeing as it hit the market in 1986 and was obsoleted soon after that, in 1989. At that time the service manuals in general were still comprehensive.

Also noticed that the "power" LED indicator near the main switch, does not work. Maybe it's not an LED, maybe it's a small incandescent bulb in there, will have to check for that.

Anyway !

Some technical  details :

- It's a 11401 main frame, good for 500MHz, sampling at a whooping 20MS/s 
- Plugin #1 : 11A32 = Vertical Amp, two channels, 400MHz. In a 500MHz mainframe I guess that will give an effective BW of what ? 350 or something ?
- Plugin #2 : 11A32 also. So a total of 4 channels.
- Plugin #3 : 11T5H = adds video trigger capabilities.

Fan is surprisingly quiet ! It's totally bearable in the long run. I was expecting MUCH worse than that to be honest !
Has RS232, GPIB and a Centronics printer port at the back.
Has of course the touch screen !    It works just fine, and it's COVID APPROVED !    You do NOT have to touch the screen surface to activate / operate it ! You only have to bring your finger close to the screen, like 5+ mm above the surface, and it will register !  It uses the same principle as the old HP touch screen computer from the '70's IIRC : an array / matrix of IR detectors that make a coarse "grid" of small rectangles about 15x10mm I would say. Something like that. The size of the tip of your pinky, say. In practice it works just fine.

It's HUGE, and HEAVY !!! Even my 500 series hollow state Teks fear this 11000 scope ! They would not want to be engaged in a fight with it ! 
It's deeper than a 500, does not fit on the bench as you can see. It's much wider than a 500 too !

The user interface is so out of this world that I really do need to read the manual to operate this thing and test it properly. It's so exotic and weird... I LOVE THIS THING!!! 

I love that video Trigger plugin too ! Has a cool 4 digit full dot matrix display, next to a little rotary encoder with a very, very firm detent, makes a big CLICK CLICK CLICK, super cool. When I fiddle with it and I watch the display next to it showing various information, it looks very much like I am fiddling with the auto-pilot controls on a modern aircraft, love it ! 

Oh boy I can't  believe I managed to grab one of these scopes, fully working, in tip top shape, with all its plugins... for free, not even shipping, delivered to my door literally ! 

A very nice and special addition to my Tek "collection" for sure !!

Hell might use it when I can ... for repetitive signals should be fine. From what I understand, it was mostly designed as a special purpose scope to help in digital circuit design, to help check and characterize with great accuracy the edges of fast digital signals. Very accurate time base, 10 bit resolution, fancy math capabilities... it's limited in what it can do, but it does it very well. Well that's my understanding of this beast so far ! 

Did I mention that I love it to bits, that I am super excited ?!   

2H30 AM... attaching a few pics and off to bed !   

[Vince]

...

[xrunner]

GOOD TEA day for me !!!! 

A friend surprised me a paid a quick visit to give me some electronic stuff he salvaged from work. He works ofr "Orange" , ex "France Telecom", the main phone operator in Frog land. He is out all day on the field maintaining the phone and Internet infrastructrure. Every time he tosses some defective and/or obsolete piece of gear, he bring it to instead of tossing it in the trash as they normally do ! 

So today he brought a phone test set, very orange as you can see, to test analog lines. Very rugged and very much water proof.

Ah yes the pheuns must be repaired!

[Cerebus]

That's familiar but I can't quite remember the manufacturer's name but it's a line processor card from one of those rack mount boxes that takes a selection of line processor cards from the front and then the physical line interfaces fit into the corresponding slot in the rear. Circa 2000 era, designed to mostly work with PDH circuits, overlaying frame relay and similar services onto them, or terminating DSL from PDH lines, perhaps some relatively low speed SDH services from STM1 up to perhaps STM4. Overpriced, underperforming and troublesome from what I remember.

[Kosmic]

A little online research revealed this article from ELEKTOR; I actually recognized the diagram from a discussion I read some while back on another random trawl. More research found a number of fleaBay listings for probes based on the design, and for the ~$8 difference between it and just getting the parts I didn't have on hand, I let laziness prevail and picked the one with the most positive feedback which was also one of the least expensive.


...

Interesting, he fixed his design (60dbm). You have an url for the blue one ? I only see the old green one on ebay.

The input capacitor on the green one is not implemented correctly. So it doesn't really go higher than 500MHz.

I posted about it over here https://www.eevblog.com/forum/testgear/hi-z-probe-for-50-ohm-spectrum-analyzer/msg3220350/#msg3220350

[Vince]

Ah yes the pheuns must be repaired!

 

[tautech]

At Vince:

[Vince]

That's familiar but I can't quite remember the manufacturer's name but it's a line processor card from one of those rack mount boxes that takes a selection of line processor cards from the front and then the physical line interfaces fit into the corresponding slot in the rear. Circa 2000 era, designed to mostly work with PDH circuits, overlaying frame relay and similar services onto them, or terminating DSL from PDH lines, perhaps some relatively low speed SDH services from STM1 up to perhaps STM4. Overpriced, underperforming and troublesome from what I remember.

Oh you know this stuff ??  What you wrote is 99% Chinese to me !  However I Googled the boards and it seems you nailed it with the " ADSL Termination " idea... This is what Google found me :

Board #1, the one with the I/O section that's the least populated compared to the other board : is branded : " ALCATEL ABLT-D".

Google found me this :

"The ABLT-D is a MULTI-ADSL POTS/BVB UD LINE TERMINATION  for the 7300 ASAM product line "

So... termination thingie for ADSL thingie ! 

Board #2 : branded " ALCATEL ADLT-L ".

Google says there is one for sale on Ebay UK for 200 pounds... and on another page it says :

" The ADLT-L is a CO ADSL LTU for the 7300 ASAM product line "

So again something to do with ADSL, to be mated with this " 7300 ASAM " product line whatever that is.

So, "ADSL stuff", that's as far as my brain can understand it !

[Cerebus]

Oh, if those are Alcatel boards then they're not for the box I was thinking of, we didn't have any Alcatel kit in-house. What I was keying on was the layout of the backplane connectors, the use of i960 series processors (a comparative rarity, I still have a PGA one somewhere), and the general size of the boards. I'm sure the manufacturer and series I was think of will come back to me after a night's sleep. To be fair to myself, all that telecoms kit is much of a muchness, form follows function and all that.

[Vince]

At Vince:

Perseverance pays ! 

I lose on 99% of the good stuff, always too late, even when I reply to ads within an hour... kinda depressed that I missed these 2 incredible deals on hollow state Teks earlier this year !   
This 11000 scope balances things out, I am happy again !

[mnementh]

What's with the random splodges of glue down the sides of several of the bigger capacitors? Someone get board on the assembly line? 

Did it come with a manual? I apparently have a spare original if needed. 

David

I think it was a trend after 2010. My conspiracy theory is they are using it bring good luck after the capacitor plague


Anyway some black magic fuckery here with one annoyance. So was futzing with the desktop PC and managed to finally get a version of Linux working on it that both supports the built in Intel i219 card and the B560 chipset. Writing this post on it now...



However it uses Wayland and fractional scaling as I have a 27" 4k monitor. If you use it at integer scale factors (100% or 200%) it's too big or too small. If you use it fractional factors, some apps (see Kicad above) are blurry because the use the X backend rather than Wayland backend. GRRRRRRRRR. This works 99% ok on windows and 100% ok on macOS 

I am considering turfing the 4k monitor and just getting a 27" 1440p one. Might be less of a pain in the butt 

I've seen silicone used (supposedly, according to service docs) to reduce audio noise (singing) in capacitors filtering output on SMPS in monitors. No, I have no idea why they didn't just design with a better switching frequency.

I'm using a 32-in 1440p monitor with VA panel. The recent generations are getting much better; rivaling IPS and faster frame rates too.

mnem
 

[mnementh]

On a more positive note I am at 49 lbs weight loss.

22 kg? Bloody hell, well done.

Exactly what I was thinking... Word for word.

mnem
 

[mnementh]

A little online research revealed this article from ELEKTOR; I actually recognized the diagram from a discussion I read some while back on another random trawl. More research found a number of fleaBay listings for probes based on the design, and for the ~$8 difference between it and just getting the parts I didn't have on hand, I let laziness prevail and picked the one with the most positive feedback which was also one of the least expensive.

   ...

Interesting, he fixed his design (60dbm). You have an url for the blue one ? I only see the old green one on ebay. The input capacitor on the green one is not implemented correctly. So it doesn't really go higher than 500MHz.

I posted about it over here https://www.eevblog.com/forum/testgear/hi-z-probe-for-50-ohm-spectrum-analyzer/msg3220350/#msg3220350

I was surprised myself to see the board not only different color but redesigned, tho I didn't look too deeply into the differences.

The listing I bought from is here: https://www.ebay.com/itm/332393676082

It still shows the green ones. This one is decent enough quality 1.6mm FR4. 

mnem
*toddles over to scope out that post*

[cyclin_al]

...it takes 1.5kWh per half turn, the same as to boil 8 cups of water!

That sounded wrong given that my 3kW kettle boils ~1.5 litres in under 3 minutes, so 1 ran the numbers. 1 cup = 220ml = 220gm, x8 -> 1.76 kg. Specific heat capacity of water 4.2 kJ/kgK, heating from 20ºC->100ºC -> ~590 kJ -> 0.165 kWh. Some one's off by a factor of 10.

Obviously, that means the ambient temperature is not 20 deg C!

I can only conclude that you do not work at the Tourism Commission for the UK.
Thanks for the warning, I will pack something much warmer than a winter parka and toque.
 

You'll need something more than a winter parka and toque, you'll need something to cope with exotic matter.

1.5 kWh = 5.4 MJ, 1.76kg of water has a heat capacity of 1.76kg x 4200 J/kgK -> 7392 J/K. Heat of fusion of 1.76kg water is 587,840 J so that leaves us 4.81 MJ. 4.8 MJ / 7392 J -> \$ \Delta t \$ of 651 K.

Boiling point of water is 373 K so with that \$ \Delta t \$ we end up at -278ºK, which leaves us in a different universe where negative thermodynamic temperatures are possible.

Good thing I wrote "much warmer than" to show that some brain cells were still functional.  At 1am local time, I was much more focused on the humour aspect than the actual numbers 

[mansaxel]

That's familiar but I can't quite remember the manufacturer's name but it's a line processor card from one of those rack mount boxes that takes a selection of line processor cards from the front and then the physical line interfaces fit into the corresponding slot in the rear. Circa 2000 era, designed to mostly work with PDH circuits, overlaying frame relay and similar services onto them, or terminating DSL from PDH lines, perhaps some relatively low speed SDH services from STM1 up to perhaps STM4. Overpriced, underperforming and troublesome from what I remember.

Alcatel? France Telecom have a long story of not recognising the presence of an international competition to gear made in France..

I took one look at it and came to the same conclusion as you wrt the purpose of the board -- linecard for an obsolete telecoms transmission technology.

There's probably more bank fraud and money laundering being made per day over obsolete comms gear as such, between legacy systems, than most small countries can make in GDP per year.


This, however, pales in comparison what the telcos make in supporting it.
 

[tautech]

A little online research revealed this article from ELEKTOR; I actually recognized the diagram from a discussion I read some while back on another random trawl. More research found a number of fleaBay listings for probes based on the design, and for the ~$8 difference between it and just getting the parts I didn't have on hand, I let laziness prevail and picked the one with the most positive feedback which was also one of the least expensive.

   ...

Interesting, he fixed his design (60dbm). You have an url for the blue one ? I only see the old green one on ebay. The input capacitor on the green one is not implemented correctly. So it doesn't really go higher than 500MHz.

I posted about it over here https://www.eevblog.com/forum/testgear/hi-z-probe-for-50-ohm-spectrum-analyzer/msg3220350/#msg3220350

I was surprised myself to see the board not only different color but redesigned, tho I didn't look too deeply into the differences.

The listing I bought from is here: https://www.ebay.com/itm/332393676082

It still shows the green ones. This one is decent enough quality 1.6mm FR4. 

mnem
*toddles over to scope out that post*

Why didn't you just power it with USB mnem ?

[Zoli]

What's with the random splodges of glue down the sides of several of the bigger capacitors? Someone get board on the assembly line? 

Did it come with a manual? I apparently have a spare original if needed. 

David

I think it was a trend after 2010. My conspiracy theory is they are using it bring good luck after the capacitor plague
...

Mine had been made around 2000(too lazy to look it up exactly), and has the RTV/silicone fix(see here). The reason AFAIK is anti-vibration, as I've seen some other stuff(PC PSU, early(<2004) plasmatv's etc.).

[Andrew_Debbie]

Those gold plated pins & sockets look a lot like those found in 1980's Tek oscilloscopes, in fact I was looking at some earlier in the Tek 214* at work.

You have missed the Telfon? insulated pin & blue socket, to the right of the row of pins/sockets, HP/Dymec seem to use the same in some voltmeters, the 2402A has quite a few and I've yet to identify something to use to make jumper wires top go with the standard extender board (of course the original custom extenders are complete unobtainium  ).

The blue pins and sockets appear to only  locate the board and provide mechanical stabilty.     I'll double check next time I have the cards out.   

The other sockets allow for some misaligment of the pins.    The blue pins and sockets do not.


'Dupont' leads might work.  I don't have enough on hand to try.   I just ordered  several types from Amazon.

[bd139]

Later than I'd hoped, but here at last are the photos of the TTi SM630, it is just as if it had just been made at the factory, inside and out. Needs a little cleaning on the case, but thats to be expected. Going by the date code on the chips, and there are many of them, this was made in 1994 or later. The backup battery looks new and judging by the soldering on the back of the board, those three connections are all new and shiny, unlike the others which have that aged dull appearance.

I have run the scope through its paces using the Heathkit IG4505 calibrator, and it hits the right frequency and periods as per calibrator, so no wuckers there. Multimeter part also meets the requirements, agreeing with both the AD584-M but also the current ratings against the Mastech MS7221 calibrator and the frequency counter is pretty much on the money too.

The backlight, as already mentioned in a previous post, is useless. It is one of those reflective types with a white panel and maybe unobtainuim, will speak to TTi about this aspect. If it is as I suspect, unrepairable, then it really becomes a SM620 which never had a backlight anyway.

I'll have to do the function generator in another post.

Nice bit of kit that. I want one now 

The display is an Optrex DMF697. EL backlight. Looks like you can get replacements for about £8 on eBay 

https://www.ebay.co.uk/itm/Optrex-DMF697-equivalent-part-Graphic-LCD-Display-Module-128x64-/284139570122