Test Equipment Anonymous (TEA) group therapy thread

[Robert763]

Got a gift of TE from SWMBO today (apparently we have been married for 18 years). A Peak Atlas ESR70 Gold.
It's very nice and fills a gap in my toolset. It's easy to use and seems accurate based on checks with some new and known bad capacitors. Interestingly the capacitors I recently removed from a HP 8920A PSU because they were leaking electrolyte checked OK for ESR and were just a bit below nominal capacitance. One to watch out for as the leakage wasn't apparent until I removed the capacitors from the board.
One interesting featute of the ESR70 is the leads are not polarised.

Robert.

[TERRA Operative]

Test equipment, from the lady? She's a keeper!

[Vince]

Wow that's one cool wife. 18 years is nothing special yet you got a nice TE... so when comes the 20 year mark, which is a nice round number, you may well get a brand new HP 20GHz spectrum analyzer with track gen. She is probably already saving for that.

Yes nice wife it is indeed....

[TERRA Operative]

My wife just tolerates me.

[Vince]

Yeah but that's a good beginning ! 

I think you have not known your wife for as long as Rob's wife.

I mean your situation might improve over the coming years and maybe in 10 year she too, will start buying you TE as well ! 

[tautech]

Wow that's one cool wife. 18 years is nothing special yet you got a nice TE... so when comes the 20 year mark, which is a nice round number, you may well get a brand new HP 20GHz spectrum analyzer with track gen. She is probably already saving for that.

Yes nice wife it is indeed....

Have you any idea of the cost of a new 20+GHz HPAK analyser ?

[Vince]

Yes, I do  !

Have you any idea of what humour looks like ?! 

That's what it looked like 

[tautech]

Yes, I do  !

Have you any idea of what humour looks like ?! 

That's what it looked like 

Oh humor, well isn't it funny a fully optioned SSA5083A is under 1/2 the cost of an equivalent HPAK analyzer. 

[EggertEnjoyer123]

I'd be happy if my (nonexistent) wife gave me a HP 8566B.

Spectrum analyzer technology hasn't improved much (the architecture is still the same, you have a YIG preselector, harmonic mixer, phase locked YIG synthesizer for the LO, second and third converters, crystal filters, etc). It's not like oscilloscopes, where an analog scope from the past has way less features than a digital scope now. Analog and digital scopes are also different in how the frontend is handled, unlike with spectrum analyzers, which have the same RF components. The only thing different with modern spectrum analyzers is that they can get a better RBW using digital processing like FFT, which older spectrum analyzers didn't have. The improved RBW isn't that great though, since you are limited by the phase noise of the internal oscillator in how much you can zoom in. Also, modern spectrum analyzers have a preamp (though you could just buy one for an old spectrum analyzer too).

You can get an HP 8566B (22 GHz) off ebay for the same price as a 3.2 GHz Siglent spectrum analyzer. I know which one I'd choose.

[Vgkid]

Just would have been curious about the internals and the firmware of a 1693 digibridge compared to the smaller versions, but no way, it went for $945: https://www.ebay.com/itm/295866213443

And the seller had only a screen with 88888 on display, which is a part of the startup test, but claims fully working...

So, if someone has one of these, i would be interested in a ROM image and pictures of the pcb board.

As the owner of a 1689 , and 1658 I have always wanted one of these , but even broken are too much for me . My 1689 was less than half what that went for...
Now I did manage to get a users/service manual for one. Going from memory the mainboard was the same as the 1689 , but there was an add on board. That used an onboard dc/dc converter , its own cpu , and a bunch of ALU's.

[tautech]

I'd be happy if my (nonexistent) wife gave me a HP 8566B.

Spectrum analyzer technology hasn't improved much (the architecture is still the same, you have a YIG preselector, harmonic mixer, phase locked YIG synthesizer for the LO, second and third converters, crystal filters, etc). It's not like oscilloscopes, where an analog scope from the past has way less features than a digital scope now. Analog and digital scopes are also different in how the frontend is handled, unlike with spectrum analyzers, which have the same RF components. The only thing different with modern spectrum analyzers is that they can get a better RBW using digital processing like FFT, which older spectrum analyzers didn't have. The improved RBW isn't that great though, since you are limited by the phase noise of the internal oscillator in how much you can zoom in. Also, modern spectrum analyzers have a preamp (though you could just buy one for an old spectrum analyzer too).

You can get an HP 8566B (22 GHz) off ebay for the same price as a 3.2 GHz Siglent spectrum analyzer. I know which one I'd choose.

Ya think.
Maybe you should borrow a new analyzer sometime and explore the capabilities these new things have, not to mention the 10 or 12" touch and mouse capable LCD displays, webservers for remote control or the compact low weight 2 finger lifts they are rather than the 2 man lift hernia creators old stuff was.

[wkb]

I'd be happy if my (nonexistent) wife gave me a HP 8566B.

Spectrum analyzer technology hasn't improved much (the architecture is still the same, you have a YIG preselector, harmonic mixer, phase locked YIG synthesizer for the LO, second and third converters, crystal filters, etc). It's not like oscilloscopes, where an analog scope from the past has way less features than a digital scope now. Analog and digital scopes are also different in how the frontend is handled, unlike with spectrum analyzers, which have the same RF components. The only thing different with modern spectrum analyzers is that they can get a better RBW using digital processing like FFT, which older spectrum analyzers didn't have. The improved RBW isn't that great though, since you are limited by the phase noise of the internal oscillator in how much you can zoom in. Also, modern spectrum analyzers have a preamp (though you could just buy one for an old spectrum analyzer too).

You can get an HP 8566B (22 GHz) off ebay for the same price as a 3.2 GHz Siglent spectrum analyzer. I know which one I'd choose.

Ya think.
Maybe you should borrow a new analyzer sometime and explore the capabilities these new things have, not to mention the 10 or 12" touch and mouse capable LCD displays, webservers for remote control or the compact low weight 2 finger lifts they are rather than the 2 man lift hernia creators old stuff was.

Of course new SA have capabilities that the SA of old could only dream of. But separating eye candy (GUI, touch screens, whatnot) from high grade RF design and build quality is, IMHO, very important.

[tggzzz]

I'd be happy if my (nonexistent) wife gave me a HP 8566B.

Spectrum analyzer technology hasn't improved much (the architecture is still the same, you have a YIG preselector, harmonic mixer, phase locked YIG synthesizer for the LO, second and third converters, crystal filters, etc). It's not like oscilloscopes, where an analog scope from the past has way less features than a digital scope now. Analog and digital scopes are also different in how the frontend is handled, unlike with spectrum analyzers, which have the same RF components. The only thing different with modern spectrum analyzers is that they can get a better RBW using digital processing like FFT, which older spectrum analyzers didn't have. The improved RBW isn't that great though, since you are limited by the phase noise of the internal oscillator in how much you can zoom in. Also, modern spectrum analyzers have a preamp (though you could just buy one for an old spectrum analyzer too).

You can get an HP 8566B (22 GHz) off ebay for the same price as a 3.2 GHz Siglent spectrum analyzer. I know which one I'd choose.

Ya think.
Maybe you should borrow a new analyzer sometime and explore the capabilities these new things have, not to mention the 10 or 12" touch and mouse capable LCD displays, webservers for remote control or the compact low weight 2 finger lifts they are rather than the 2 man lift hernia creators old stuff was.

Of course new SA have capabilities that the SA of old could only dream of. But separating eye candy (GUI, touch screens, whatnot) from high grade RF design and build quality is, IMHO, very important.

True, and overlooked in many "reviews" of many types of equipment.

Yootoob reviews are mostly junk along the lines of "I opened the pretty box, turned it on, and it worked". Too many "professional" reviews are not much better.

[Robert763]

Wow that's one cool wife. 18 years is nothing special yet you got a nice TE... so when comes the 20 year mark, which is a nice round number, you may well get a brand new HP 20GHz spectrum analyzer with track gen. She is probably already saving for that.

Yes nice wife it is indeed....

Have you any idea of the cost of a new 20+GHz HPAK analyser ?

She got me a R&S  spectrum analyser with TG for my birthday last year.. She has also found very good books without input from me. She is Dr SWMBO but it's education / language not STEM.

[Robert763]

I'd be happy if my (nonexistent) wife gave me a HP 8566B.

Spectrum analyzer technology hasn't improved much (the architecture is still the same, you have a YIG preselector, harmonic mixer, phase locked YIG synthesizer for the LO, second and third converters, crystal filters, etc). It's not like oscilloscopes, where an analog scope from the past has way less features than a digital scope now. Analog and digital scopes are also different in how the frontend is handled, unlike with spectrum analyzers, which have the same RF components. The only thing different with modern spectrum analyzers is that they can get a better RBW using digital processing like FFT, which older spectrum analyzers didn't have. The improved RBW isn't that great though, since you are limited by the phase noise of the internal oscillator in how much you can zoom in. Also, modern spectrum analyzers have a preamp (though you could just buy one for an old spectrum analyzer too).

You can get an HP 8566B (22 GHz) off ebay for the same price as a 3.2 GHz Siglent spectrum analyzer. I know which one I'd choose.

Sorry you are wrong. Most modern design spectrum analysers use significantly different architecture from the classic lie the 8566B. There are two main changes both based on digital signal processing (DSP) techniques:
1/ A DSP IF system including filters This was the first big improvement but is fairly transparent to the user.
2/ Real Time spectrum analysers. This is where the RF is directly digitised, normally as I/Q (quadrature) signals. All the processing is done by software.
There are practical problems with Real Time spectrum analysers when trying to make directly comparable measurements with different instruments.
I'm a memeber of a committee that sets (among other things) EMI emissions standards for aircraft equipment. With the conventional analyser you just specified sweep rate, bandwiths and detector type to get the same result (and most labs used 8566A/B anyway) but with modern ones there are too many software subtleties and tweaks that could be abused. And diffrent makes use differnt terminology....

Robert.

[EggertEnjoyer123]

I’m talking about the RF frontend. The part which converts the 20GHz or whatever signal to IF is the same still. The IF is handled digitally but I said that in my comment. In any case a 8566B can be bought for a kilobuck and a new 22GHz spectrum analyzer would probably set you back $50,000 so it’s an easy choice. I’ve used the new ones at work and the UI is better, but it’s not something I’d pay 50 thousand dollars for as a hobbyist.

I guess I was wrong about crystal filters still being used. Have to find someone who’s willing to let me take apart one of those newer spectrum analyzers. I also didn’t think of real time ones, but those are probably even more expensive and I definitely can’t afford them.

Edit: You can buy a 18ghz real time SA from Tek but it’ll set you back $29,000

[Robert763]

The front end is different. The RF is directly sampled for the baseband in the new real time instruments. There is no swept local oscillator. Even the 8566A/B is only a 1.5 GHz analyser (swept LO) with a harmonic mixer down converter. The 8568A/B is the same instrument without the down converter.

Edit: the 8566A/B cost far more (even without allowing for inflation) than a modern 20GHz RT SA. The 8568A was $27,800 and the 8566A $47,500 in 1980. (about $175,000 today).

[m k]

Did you even look at the pics he posted ?! 

The second screen shot clearly lists the RS-423....

Dang, actually I did, long enough that clean glasses wouldn't do any change.
It's cölear again that eyes don't see.
Expectation it can't be either so who knows where that image went.

I think it actually was expectation.
My understanding was probably concentrated to protocols and signaling was put aside.

How that actually happened is of course unknown but it's interesting since reading human is seeing words, not letters.
Here it is a bit special since there are also other markings than letters.
Maybe my understanding took only parts before opening bracket, put the rest to addendum info section and used already known old info as a main thing.
I've also not checked the picture again, bracket is from the memory, maybe there aren't any.

An anecdote,
the phaonmneal pweor of the hmuan mnid, aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it dseno't mtaetr in waht oerdr the ltteres in a wrod are, the olny iproamtnt tihng is taht the frsit and lsat ltteer be in the rghit pclae. The rset can be a taotl mses and you can sitll raed it whotuit a pboerlm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.

[m k]

Wow that's one cool wife. 18 years is nothing special yet you got a nice TE... so when comes the 20 year mark, which is a nice round number, you may well get a brand new HP 20GHz spectrum analyzer with track gen. She is probably already saving for that.

Yes nice wife it is indeed....

Maybe it's something different.
Maybe Robert has had few hints how to behave when special 20 arrives.

[Robert763]

Well for 18th SWMBO got a new carbon fibre bow. For violin not arrows. It cost ten times the ESR 70 Gold....
Last year I got a small mill and she got a couple of CD's. Swings and roundabouts.

Robert.

[Ice-Tea]

HP6032A with the miraculous capacity of turning the lights off on this floor with its power switch on the bench. Opening it up reveals quite a lot of corrosion. Some of that is without a doubt from being stored at you-know-where but not all. The fan is seized hard and the previous owner apparently just disconnected the fan. He had a point, I suppose, with all that alluminum it will probably run fine at lower power outputs and will take time to overheat full tilt.



So, what could poss... Oh, right 



Browsing the schematics I get this strange feeling in my stomach and sure enough:



Congratulations, mr. Ice-tea, you just booked an additional hour in your callendar to get this thing back on the road. But... at least it runs again!

[TERRA Operative]

That looks uncannily lile my 6035A inside and out for some strange reason, almost like they were related models or something.

[wkb]

Ah, yeah, the RIFA pest. Recently harvested a bunch out of my new-to-me HP8595E spectrum analyser. For good measure I replaced the Schaffner IEC power inlet/mains filter/fuseholder combo while I was there. For the RIFA 100nF I need to source form fitting X caps, space is tight under the Schaffner. For now the HP does without, better this way than waiting for them to pop.

3x RIFA lurking under the Schaffner mains inlet/filter.
Some innocent WIMA's as well.

[Vince]

Hello !

Some action this morning, on my Tek 2465 scope.

I received Factory's envelope, containing that HV 3.3uF cap I needed to complete the PSU recap.
So I did that.

I also received my replacement / used brushless fan motor from QService i Greece, so I installed that as well.
Removing the old one was a bit scary given it's got no less than 13 pins, arranged in a circle, and I have no desoldering gun, and half the pins are soldered on the top side of the PCB, underneath the motor, so invisible and inaccessible, great.
Knowing that the motor was dead was a relief as I didn't have to care about it, but still.

Anyway, solder wick it was. I struggled a bit to pry the PCB off of the motor's butt, but not too much, and no damage, it went OK I would say.
Then I solder the replacement motor in place, then try to fit it in its plastic housing to make sure everything is fine and... no, everything was not fine. It would not fit 10% right... then realized the old motor was fitted with a plastic spacer over it's pins, which of course I didn't pay attention to and didn't swap.
So... the freshly replacement motor had to come off already !    Lucky me, desoldering it was a breeze, the hidden solder joint gave zero trouble and the PCB came right off, zero prying necessary. A miracle.

So fitter the spacer, soldered it back, then gave it's control board some power to see if that motor would work better than the old one... well, azt first it seemed so : the motor would be able to develop a lot more torque than the old one, for sure enough torque to spin the plastic fan. So I was relieved.... but not for long : after just a minute, the motor would "lock up" : motor had stopped, and if I try to turn it by hand, it actively resists being turn either direction. The windings are energized in such a way that it does so.. weird. I thought OK maybe one or more of the 4 transistors I put to drive the windings, gave up the ghost. I didn't have proper replacement for the original DIP package transistor array, remember, so I used TO-30 2N2222A instead.
So I test them... still good, strange. Put them back in place, power the motor again... it recovered, it turns again now... 

So, I fit the motor assembly to the PSU, fit the PSU back into the scope, and fit the fan onto the motor shaft... the collet is broken in two, and I am missing the nut to tighten the collet as well, but somehow if I just shove the broken collet inside the fan, and fit that to the shaft... there is enough friction for the fan to grab the shaft, miracle.

So I power up the scope... and yes, miracle, the fan spins now, which it did not before !
... but again... 2 minutes later, same crap happens : motor "locks up" and fan can't spin anymore 

Well, moving on, I checked all voltages and ripple as per the manual, adjusted the 10V ref voltage spot-on to 10.000V, which is not that easy as the trimmer they provide on the PSU is extremely touchy. Anyway, once the 10V ref was adjusted, all other rails were fine, and ripple next to zero, so much so that I had to double-check using a different DMM, but they agreed.

Result ? The PSU did not explode, and the scope still works, incredible ! 
However I notice the readout is still unstable/wavy at power up, and calms down after a few minutes. Bummer 
But since the PSU is recapped and ripple is fine... at least we know this instability is not due to the PSU. So it must be due to some other electrolytic cap, somewhere else in the scope, wherever the signal path for the read-out is located. I guess it spans both the CPU board and the main analog board.
I am eager to figure it out because my 2467B suffer from exactly the same problem (it too has been recapped), so I could kill two birds with one stone.

So in conclusion :

- The recap is done and went well

- Fan collet still need fixing. I also need to source the missing 4-40 imperial nut that's used to tighten that collet. No, we don't have these nuts in Frogland 

- Motor : Don't know whether it's bad and I lost lots of money (60 freaking Euros !), or if the problem is due to my inappropriate 2N2222's. So I first must identify and buy some better suited transistors, and see if that improved things. If it does not... the motor is probably defective and not only did I lose lots of money, but I am also back to square one.  My gut feeling is that the motor is bad, but fitting new transistors is cheap and easy so I will do that first and see what happens. If the motor is indeed bad then I will try to get a refund and then I will have to start a whole new project / journey to replace this motor with a different one that's readily available brand new, cheap, and of course will have to design a whole new control board to go with it. Fun. Well not really. But one thing at a time.. first the transistors, with some luck...



Something else :  the cap was also delivered with ... a foot I was missing for one of my two old HP ratio meters, thanks Dave ! 
However I failed to fit it... I not missing just the foot but also... the screw securing it to the chassis ! 
So I need to look up in the manual, hopefully there is a mechanical parts list that will tell me the diameter and pitch I need.
So I need an imperial nut for the scope, and an imperial screw for the HP foot. Pfff.....


So all in all, making progress, not complaining but... still a long way to go to get that freaking fan working 

[TERRA Operative]

I wonder if the vibration of the fan motor is playing games with the contact between the socket and the dangly transistors?