Test Equipment Anonymous (TEA) group therapy thread

[tggzzz]

I would indeed test that but I am stuck with a ~10nS rise time output here. I suppose it's a good excuse to build a Jim Williams pulse generator...

Don't bother; there are much easier alternatives with modern logic.

A 10p 74lvc1g1* will be more than adequate. If you add a 143ohm series resistor, the combination will have a 150ohm output. Put three of those in parallel and you will have a very clean 50ohm generator that operates with a Vcc between 1.5 and 5V. I constructed one, and an indirect measurement gives the risetime of ~600ps; someone else measured it with a modern scope and it looked like it was 300ps or less. The only trick is to make sure they are very well decoupled!

Slightly more "exotic" is the output from an Analog Devices comparator; the risetime is ~50ps. There's another thread on this site, and the author is selling them.

I must admit when I cracked it open, the timebase did leave me a bit "whaaat?" in this case because the board wobbles when you turn the knob. I will probably leave it be for now and just buy another 1740A to keep in the cupboard just in case

The wobbles don't worry me too much, and should be minimal if the supporting "doughnuts" are positioned correctly. Remember IBM PC ISA boards!

Alas no 485 has ever come my way for a price I'd pay or I'd snap it up instantly ... still slightly regretting selling my 475 which was in mint condition a couple of years ago. That was second best to the 485.

I paid £50 for mine, but I had to diagnose and recap the PSU, including the 15V tant caps on a 13V line that isn't shown on the PSU board!

On a second one, now sold for a bargain £150, the resistor inside an IC had failed, so I soldered a SMD resistor across nearby terminals. Pleasingly the only visible issue was that the risetime was 1.25ns instead of 1ns; yes I did inform the buyer!

[Berni]

Especially when it still has its original badge on it!

So far the 3 pieces of gear i got with the classic badge came with the original one. Others have the HP logo be part of the front panel design or is new enough to say Agilent on it.

For some time Agilent made there own version of the old HP badge that's the same size and fits in to the same two hole badge pattern except its a silver Agilent logo on black background, more in tune with the design of the newer black and silver HP badges that are found on the newer test equipment before they became Agilent. Tho those look like they belong on a inkjet printer rather than test gear.

Too bad they now just print the logo on the front panel like everyone else.

[bd139]

The first rule of TEA - never sell!  Selling eventually leads to regret.

I suffer this regularly but will never learn!

Well it is easier to sell modern cheap Chinese test gear that didn't really perform up to your expectations anyway, so you are happy to never see it again.

I'm going to be honest: I recently bought a DS1054Z and I'm not totally enamored with it. If I was going to sit and do this for a living then I'd be happy with it as it's a tool but as I said earlier, it lacks soul. Plus it sounds like an angry wasp. I'm not sure I'm too happy about the noise on it either. One thing I genuinely miss is the delay sweep feature which there isn't a good solution to on the Rigol. Basically I say "hey that's a bit fuzzy", set the delay sweep to lowest time, pop the timebase selector and there you go, it's oscillating! Rigol: bring on knob-fuckery and then you have to go back to where you were which is difficult. That leads to be just hitting "auto" followed by two channels showing 50Hz because the probes are just dangling and disappointment.

Its probably the hardest for the old name brand equipment. Especially when its in particularly good cosmetic condition and fully working, but at the same time you just bought a new piece of equipment that does the same job with better performance at 1/4 the size and weight.... but the old boatanchor has that lovely shiny chrome HP badge on it.

My wife LOVES the DS1054Z because of the size. My bench is in the corner of the living room. She's not going to like the HP when she gets back after the weekend. I actually picked the 1740A up on the way back from dropping her off

I would indeed test that but I am stuck with a ~10nS rise time output here. I suppose it's a good excuse to build a Jim Williams pulse generator...

Don't bother; there are much easier alternatives with modern logic.

A 10p 74lvc1g1* will be more than adequate. If you add a 143ohm series resistor, the combination will have a 150ohm output. Put three of those in parallel and you will have a very clean 50ohm generator that operates with a Vcc between 1.5 and 5V. I constructed one, and an indirect measurement gives the risetime of ~600ps; someone else measured it with a modern scope and it looked like it was 300ps or less. The only trick is to make sure they are very well decoupled!

Slightly more "exotic" is the output from an Analog Devices comparator; the risetime is ~50ps. There's another thread on this site, and the author is selling them.

Thanks - looks like an interesting solution and considerably less crap than having to step up the voltage to 90v or so first. Will do some research and build one. I have a couple of nice BNC connectors floating around so I can couple it directly to the scope. Might even buy a nice little box

I must admit when I cracked it open, the timebase did leave me a bit "whaaat?" in this case because the board wobbles when you turn the knob. I will probably leave it be for now and just buy another 1740A to keep in the cupboard just in case

The wobbles don't worry me too much, and should be minimal if the supporting "doughnuts" are positioned correctly. Remember IBM PC ISA boards!

I do indeed. My father was an importer of white box ISA card crap from Taiwan in the late 1980s. We had the little foam padded boxes stacked up to the ceiling in the house. On the positive side, free PC

Alas no 485 has ever come my way for a price I'd pay or I'd snap it up instantly ... still slightly regretting selling my 475 which was in mint condition a couple of years ago. That was second best to the 485.

I paid £50 for mine, but I had to diagnose and recap the PSU, including the 15V tant caps on a 13V line that isn't shown on the PSU board!

On a second one, now sold for a bargain £150, the resistor inside an IC had failed, so I soldered a SMD resistor across nearby terminals. Pleasingly the only visible issue was that the risetime was 1.25ns instead of 1ns; yes I did inform the buyer!

You had some good luck there!

Next on my list is a Tektronix 7000 series (7504 would be nice due to construction) and 7L12 if I can find one. I figured I can probably get both for less than a Rigol DSA.

Now I'm going to think about selling the DS1054Z...

[med6753]

Years ago when I wasn't active I sold several of my Heathkits plus a rare Tektronix 561S scope that I had restored. For some reason I kept the manuals so I'm reminded on a daily basis of my stupidity.  Never again! 

[bd139]

I can't stand Heathkit stuff. I used to buy it because it had decent service manuals but usually it turned out to have either been assembled by a drunk monkey or modified by a drunk monkey. Both drunk monkeys appear to have used a 100W Weller soldering gun. On top of that, the design was universally "cost efficient". As in missing things like safety features and running everything literally on the verge of exploding into pieces.

My first bench supply was a Heathkit IP-18 that a rat had used as a toilet.

[med6753]

I can't stand Heathkit stuff. I used to buy it because it had decent service manuals but usually it turned out to have either been assembled by a drunk monkey or modified by a drunk monkey. Both drunk monkeys appear to have used a 100W Weller soldering gun. On top of that, the design was universally "cost efficient". As in missing things like safety features and running everything literally on the verge of exploding into pieces.

My first bench supply was a Heathkit IP-18 that a rat had used as a toilet.

Agree to a point. I've seen some of the drunk monkey assemblies and you just want to shake your head. 

But these units were built by me and they all worked as designed. Oh well.     

[bd139]

Congratulations on not being a drunken monkey

The main thing that always concerns me about Heathkit stuff is the power supplies. They appear to be avoiding buying regulators and running a zener with a pass transistor. That's fine but not when your logic is running at 6v due to the crappy tolerance zener!

[bd139]

Well I spent the afternoon on the new HP 1740A giving the panel and controls a good clean and checking calibration.

Calibration is quite surprisingly absolutely bang on. Apart from some wear on the timebase control, the thing is cleaning up pretty well:



Cleaning tools used: Waitrose's Dettol rip off, tea towel, paint brush, Wera TX4 Torx driver. Nothing high tech

[bitseeker]

Wow, that's looking great! Now I want one.

[tggzzz]

Wow, that's looking great! Now I want one.

You can have mine - for £1000. I'll throw in postage and packing for free

[bitseeker]

 

Nevertheless, I wouldn't want to risk it getting trashed in shipment.

[bd139]

You could get a lot of bubble wrap for £200 of that. If it fell in the sea it'd still float there

[xrunner]

Well I spent the afternoon on the new HP 1740A giving the panel and controls a good clean and checking calibration.

Cleaning tools used: Waitrose's Dettol rip off, tea towel, paint brush, Wera TX4 Torx driver. Nothing high tech

Absolutely great job! Couldn't have done better myself. 

[Cubdriver]

That looks brand spankin' new!  Nice work!!   

-Pat

[VK5RC]

A beauty! Well done. 

[bd139]

I found another excuse to buy more equipment.

Decided I wanted to use the DG1022Z's frequency counter to monitor VFO drift so I'm automating it with SCPI to take a reading every 5 minutes. This turned out to be rather easy so I've spent the last hour writing a C# client library.


using (var arb = new RigolDG1000Z(new SCPIEndPoint("tcp://192.168.1.4:5555")))
{
    arb.Counter.State = RigolDG1000ZCounterState.ON;
    Thread.Sleep(1000);
    arb.Counter.State = RigolDG1000ZCounterState.OFF;
}


Doesn't do a lot yet but the above works. Working on a response parser now so I can do query commands. This can then talk to Excel via COM and add the data to a workbook.

Now the TEA excuse: I need lots of things that talk SCPI so I can write client libraries for them and make them dance like the electric puppets they are

[AF6LJ]

Jones'in for some test equipment, ham radio swap meet tomorrow.
Money in my purse....

[Mr. Scram]

Jones'in for some test equipment, ham radio swap meet tomorrow.
Money in my purse....

Money in, garbage delicious ham out.

[Cubdriver]

Jones'in for some test equipment, ham radio swap meet tomorrow.
Money in my purse....

You going to TRW, or something more local to you down south?  Either way, happy hunting!!!

-Pat

[bitseeker]

I found another excuse to buy more equipment.

Oh, didn't know you needed a source of excuses.

Now the TEA excuse: I need lots of things that talk SCPI so I can write client libraries for them and make them dance like the electric puppets they are

You're in luck. Lots of things talk SCPI. I guess we won't be seeing you for a long, long time.

[neo]

I think i found a new type of tea, got a project in the works that uses over a hundred 7400 series ICs just because i prefer to do it with basic logic. I'll tell you what it is if i ever get it working, which is a slim chance.

[Tony_G]

Well I spent the afternoon on the new HP 1740A giving the panel and controls a good clean and checking calibration.

That looks fantastic - Nice clean traces and everything - Congrats.

[alm]

Decided I wanted to use the DG1022Z's frequency counter to monitor VFO drift so I'm automating it with SCPI to take a reading every 5 minutes. This turned out to be rather easy so I've spent the last hour writing a C# client library.

If you want to implement a standard interface that will work across different equipment models, you could look at the IVI standards. You can find IVI drivers for many instruments.

[bd139]

Yeah looked at that. Decided it looked like a nested abstraction from hell. SCPI took me about 5 minutes to work out from plugging the Ethernet cable in.

TBH I've actually done something neat with the DS1054Z. I wanted to read RF power rather than Vpp so I used Linux's nc/awk/dc commands to just read Vpp off the device via SCPI. Literally a one liner.

[bitseeker]

I'll tell you what it is if i ever get it working, which is a slim chance.

Does that mean we'll never know what it is? Should we take guesses in the meantime?