Test Equipment Anonymous (TEA) group therapy thread

[Kosmic]

For low impedance probes matching is not really an issue because generally "as low a capacitance as possible" is the target and the issue of of HF and LF response balancing is less of a problem as the impedance seen by the probe input is both lower and flatter. To get a flavour of this, let's look at the impedance curve for that TPP1000 10MΩ probe:



You can see that as we get to higher frequencies the input impedance is dominated by the reactance of that 3.9pF input capacitance. The reactance of 3.9pF at 100 MHz is 408Ω, and it is no coincidence that's the region that the probe impedance is in by the time we get to 100MHz. Now, for a probe with 50Ω resistive impedance the 400Ω in parallel to that has a fraction of the effect (about 44Ω total) compared to being in parallel with 10MΩ (calculated as 407Ω for completeness sake) - a reduction of 12% compared to a reduction of 99.996%.

Honestly looking at those graphs just ruins my day.... it means basically you can't probe anything at all really... the impedance drops so quickly, and the phase shit moves so fast before after the narrow flat area, that unless you are looking at a pure sine wave... the signal you see on the scope is hardly what it is actually in the DUT, so making any "measurement" is just a joke 

You get a ballpark idea of what the signal shape might be, OK... other than that, it's just an illusion 

If your signal has any significant amount of energy in harmonics, then be it the rapidly changing impedance/amplitude, or phase shift... the signal displayed on the scope is potentially heavily distorted in all kinds of weird and wonderful ways.

That's mostly why I don't like modern high-speed stuff.... even simply looking at a freaking square wave/digital signal is day dreaming. It will never look like  a square and whatever overshoot or undershoot or dampening or any distortion anywhere on the waveform... you can't tell 100% if it is real, or if it's just the scope and probe setup that creates it for you out of the blue

It is very depressing....

Other reason for not liking high-speed is the price of probes and test gear rises exponentially with the bandwidth.

Sticking to 1GHz max, is about affordable to the hobbyist yet still plenty of stuff to do to learn and have fun with.  High-speed is just a money pit and a recipe for depression.... to me at least ! 

Some people have to do it as a job because all the gizmos are fast digital stuff nowadays.. but as far as fun goes in one's personal lab, I don't see the fun in high-speed digical stuff

Now that I have got that off my chest, let me go back to tying my power cords if I may.

You probably want an active probe then 

Performance will not depress you but the price might 

[SilverSolder]

[...]
Sticking to 1GHz max
[...]

I keep it below 30MHz most of the time!  - but considering one of those nano-VNA thingies just to take an occasional peek...

There is a whole world of weirdness even below 1MHz...   Blue pill, or red pill? 

[xrunner]

Boonton Model 42A News - The Good, The Bad, and The Ugly

I thought I had it about wrapped up. It had been working perfectly for a week and I even made a lot of progress on an LED light conversion involving a neat translucent mounting plate. Then it developed a problem. The meter would not go all the way back to zero with no rf signal applied and the power meter on (no it is not any sort of problem with the actual meter movement). The meter goes back to zero as it should if the instrument power is turned off. It also would "jump" several dB on the scale while decreasing power in small steps. Every once in a while, this problem goes away for no apparent reason and the instrument will work perfectly again. When the problem is gone, I can verify all the calibration procedure adjustments are spot-on.

One thing I found was that the chopper frequency wasn't 90 Hz but 98 Hz. It is shown as 90 Hz (and sometimes the doubled F of 180 Hz) in many places on the schematic where they show what waveforms are supposed to look like. It is set by a simple unijunction oscillator that has no adjustment. I removed the resistor that is part of the frequency control and retrofitted a pot, which allowed me to set the frequency to 90 Hz. I didn't think that was the actual problem but I wanted to get things set as the documentation stated it should be.

I replaced more items like caps and some old carbon resistors when I thought I had located the issue in an area using freeze spray. However, I didn't isolate the flaky component (or components) in the end. I've inspected each board meticulously more than once, and tested as much as I can in-place and found nothing suspicious anywhere so far.

I do not have the obsolete transistors, odd-value 1% resistors, and other such items to replace anyway. Also, both sides of the bottom board (amplifier board) that the other two boards plug into has conformal coating on it making it a pain to work on. I've spent many hours on it and it's time to move on (for now).

When I can find a parts mule that has the obsolete components like transistors I'll see if I can't get closer to finding out what is wrong, hopefully swapping boards to help isolate which one is causing this issue. It does measure the correct power full-scale on each scale so the sensor is OK. Also on the lowest scale (-50 dBm) it is a very responsive field strength meter when located anywhere near my router. Like I said above, whatever it is will disappear every so often, making this a very hard issue to fix.

However, since the sensor is fine, I decided to make an offer on a Boonton Model 4220A digital power meter. The seller listed it for $99 along with "make an offer". I offered $87 and they accepted it. It should arrive today (picture attached). After initial checkout I'll get it all cleaned up and calibrated. The guy I was going to give the Model 42A to was excited and he said he'd gladly pay me back for the 4220A meter.

The saga continues ...

[cyclin_al]

MuRuPuOrKeViSiVioHaVa

That one is easy ! It's the marking code for a SOT23 5.1V Zener diode ! Everyone knew that I thought ?! 

Nope, is that something you just made up for the sheer hell of it 

Gurrgle thinks it is a geocache of dubious content:

https://www.geocaching.com/geocache/GC4BFK4_murupuorkevisiviohava

mnem
I'd go, but I need to have my scales polished... 

Nope, that geocache description looks fine to me.  Did you not see the electronic components used to make the graphical border around the text?

At a more advanced level, did you see the Winnie-the-pooh reference in that same geocache? 



EDIT:

What have you all been drinking or smoking today? 

Nothing at all.

It is waaay tooo coooold to be able to light anything up!

You know it is too cold when you take force the kid outside for some fresh air.  I looked back, and the dog decided to watch us at the patio door from INSIDE the kitchen...

[Vince]

Boonton Model 42A News - The Good, The Bad, and The Ugly

I thought I had it about wrapped up. It had been working perfectly for a week and I even made a lot of progress on an LED light conversion involving a neat translucent mounting plate. Then it developed a problem. The meter would not go all the way back to zero with no rf signal applied and the power meter on (no it is not any sort of problem with the actual meter movement). The meter goes back to zero as it should if the instrument power is turned off. It also would "jump" several dB on the scale while decreasing power in small steps. Every once in a while, this problem goes away for no apparent reason and the instrument will work perfectly again. When the problem is gone, I can verify all the calibration procedure adjustments are spot-on.

One thing I found was that the chopper frequency wasn't 90 Hz but 98 Hz. It is shown as 90 Hz (and sometimes the doubled F of 180 Hz) in many places on the schematic where they show what waveforms are supposed to look like. It is set by a simple unijunction oscillator that has no adjustment. I removed the resistor that is part of the frequency control and retrofitted a pot, which allowed me to set the frequency to 90 Hz. I didn't think that was the actual problem but I wanted to get things set as the documentation stated it should be.

I replaced more items like caps and some old carbon resistors when I thought I had located the issue in an area using freeze spray. However, I didn't isolate the flaky component (or components) in the end. I've inspected each board meticulously more than once, and tested as much as I can in-place and found nothing suspicious anywhere so far.

I do not have the obsolete transistors, odd-value 1% resistors, and other such items to replace anyway. Also, both sides of the bottom board (amplifier board) that the other two boards plug into has conformal coating on it making it a pain to work on. I've spent many hours on it and it's time to move on (for now).

When I can find a parts mule that has the obsolete components like transistors I'll see if I can't get closer to finding out what is wrong, hopefully swapping boards to help isolate which one is causing this issue. It does measure the correct power full-scale on each scale so the sensor is OK. Also on the lowest scale (-50 dBm) it is a very responsive field strength meter when located anywhere near my router. Like I said above, whatever it is will disappear every so often, making this a very hard issue to fix.

However, since the sensor is fine, I decided to make an offer on a Boonton Model 4220A digital power meter. The seller listed it for $99 along with "make an offer". I offered $87 and they accepted it. It should arrive today (picture attached). After initial checkout I'll get it all cleaned up and calibrated. The guy I was going to give the Model 42A to was excited and he said he'd gladly pay me back for the 4220A meter.

The saga continues ...

From everything you said, looks like the meter works just fine and just has an intermittent connection somewhere.. mechanical problem not thermal, so freeze spray got you nowhere with that.
So inspect all solder joints.. which are probably all just fine. Wiggle connectors...
If not that, then your best friend is a wooden chop stick and patience. Tap and torture every component on every board until you can trigger the fault...
With some luck it will be a visible hair crack in some resistor, but if not lucky then well....  a dodgy internal bond in any discrete component or inside a transistor can... so all invisible. So tap test is the only way to go I fear.... good luck !!!   

If you are "lucky" enough that the fault, when it decides to manifest itself, stays put for a decent amount of time.... maybe you could try to probe around and narrow it down to a more specific part of the instrument... then you could concentrate your tapping efforts in that area...

[tautech]

Boonton Model 42A News - The Good, The Bad, and The Ugly

I thought I had it about wrapped up. It had been working perfectly for a week and I even made a lot of progress on an LED light conversion involving a neat translucent mounting plate. Then it developed a problem. The meter would not go all the way back to zero with no rf signal applied and the power meter on (no it is not any sort of problem with the actual meter movement). The meter goes back to zero as it should if the instrument power is turned off. It also would "jump" several dB on the scale while decreasing power in small steps. Every once in a while, this problem goes away for no apparent reason and the instrument will work perfectly again. When the problem is gone, I can verify all the calibration procedure adjustments are spot-on.

One thing I found was that the chopper frequency wasn't 90 Hz but 98 Hz. It is shown as 90 Hz (and sometimes the doubled F of 180 Hz) in many places on the schematic where they show what waveforms are supposed to look like. It is set by a simple unijunction oscillator that has no adjustment.

Which one....I might have it.

[Vince]

For low impedance probes matching is not really an issue because generally "as low a capacitance as possible" is the target and the issue of of HF and LF response balancing is less of a problem as the impedance seen by the probe input is both lower and flatter. To get a flavour of this, let's look at the impedance curve for that TPP1000 10MΩ probe:



You can see that as we get to higher frequencies the input impedance is dominated by the reactance of that 3.9pF input capacitance. The reactance of 3.9pF at 100 MHz is 408Ω, and it is no coincidence that's the region that the probe impedance is in by the time we get to 100MHz. Now, for a probe with 50Ω resistive impedance the 400Ω in parallel to that has a fraction of the effect (about 44Ω total) compared to being in parallel with 10MΩ (calculated as 407Ω for completeness sake) - a reduction of 12% compared to a reduction of 99.996%.

Honestly looking at those graphs just ruins my day.... it means basically you can't probe anything at all really... the impedance drops so quickly, and the phase shit moves so fast before after the narrow flat area, that unless you are looking at a pure sine wave... the signal you see on the scope is hardly what it is actually in the DUT, so making any "measurement" is just a joke 

You get a ballpark idea of what the signal shape might be, OK... other than that, it's just an illusion 

If your signal has any significant amount of energy in harmonics, then be it the rapidly changing impedance/amplitude, or phase shift... the signal displayed on the scope is potentially heavily distorted in all kinds of weird and wonderful ways.

That's mostly why I don't like modern high-speed stuff.... even simply looking at a freaking square wave/digital signal is day dreaming. It will never look like  a square and whatever overshoot or undershoot or dampening or any distortion anywhere on the waveform... you can't tell 100% if it is real, or if it's just the scope and probe setup that creates it for you out of the blue

It is very depressing....

Other reason for not liking high-speed is the price of probes and test gear rises exponentially with the bandwidth.

Sticking to 1GHz max, is about affordable to the hobbyist yet still plenty of stuff to do to learn and have fun with.  High-speed is just a money pit and a recipe for depression.... to me at least ! 

Some people have to do it as a job because all the gizmos are fast digital stuff nowadays.. but as far as fun goes in one's personal lab, I don't see the fun in high-speed digical stuff

Now that I have got that off my chest, let me go back to tying my power cords if I may.

You probably want an active probe then 

Performance will not depress you but the price might 

Yeah they are crazy expensive, but well, in some time, who knows.. once I have all the more basic stuff acquired/ sorted and I start venturing into more fancy and exotic TE... who knows what I might want to acquire !

However, IIRC from a video about active probes a few years back (might well have been that old Tek video, can't remember), even mega expensive active probes aren't magical. Severe limitations and very fragile, no ? ISTR video said they have a very narrow dynamic range, like, ridiculously low, 5 Volts or something, which is ridiculous. Good enough for 3.3 logic stuff I guess, but that's it... unless you add an external atenuator before feeding the signal to the probe, I guess...
So the probe would have an internal mechanism to provide a DC offset voltage so you could shift the input range to where you need it to be for your particular circuit. So say the probe has a 5V dynamic range and you want to measure a signal that evolves between 30 and 35V.. you would tweak the probe to add a 30V offset to it.... and pray that your signal never goes outside that window or IIRC it can get damaged ! 

So.. expensive, yet very limited and cumbersome to use ! 

But well, there is no magic stick so if that's the best science can give to the engineers in the lab... you gotta make do with it eh ! 

[xrunner]

From everything you said, looks like the meter works just fine and just has an intermittent connection somewhere.. mechanical problem not thermal, so freeze spray got you nowhere with that.
So inspect all solder joints.. which are probably all just fine. Wiggle connectors...
If not that, then your best friend is a wooden chop stick and patience. Tap and torture every component on every board until you can trigger the fault...
With some luck it will be a visible hair crack in some resistor, but if not lucky then well....  a dodgy internal bond in any discrete component or inside a transistor can... so all invisible. So tap test is the only way to go I fear.... good luck !!!   

If you are "lucky" enough that the fault, when it decides to manifest itself, stays put for a decent amount of time.... maybe you could try to probe around and narrow it down to a more specific part of the instrument... then you could concentrate your tapping efforts in that area...

I've tried all of that Vince. But I may get back to it in a few weeks after I finish checking out the Boonton Model 4220A. It's on right now and looking good so far. Pics tomorrow ...

Oh and I got to show you the packing material they put it in. This was a first for me. 

Which one....I might have it.

Thanks for offering but I don't need anything because as I said I put in a pot and now I can set the frequency to 90 Hz.

[Vince]

Phew, done with the power cords ! 

The 2 pronged ones were a nightmare to work with... they were super stiff, and being flat rather than round, didn't hep roll them nicely. Had to straighten them first, then roll them very slowly, to keep that flat jacket from twisting and make a mess of the job.
They are so stiff, I could swear they were using rigid / solid core wires... but I cut one off to check and... no, stranded wires they sport ! Weird... so it must be the insulating jacket itself that's SO stiff ?! Boy, it's quite something. Cheap PVC but still ! Maybe after many years the PVC hardens ?!
Anyway, was a chore but I think the result is acceptable, and an order of magnitude better than it was before, for sure... so happy enough.

Went with the zip ties in the end. Cheaper than I thought. H/W store had plenty to choose from. Bought a plastic jar full of them, an assortment of 3 sizes, 500 pieces in all, for only 7+ Euros. Not complaining.

While at the store I also bought a wooden 3 drawer unit, to store and sort the cables. No luck, I under estimated the volume all these cords would necessitate... so for now it's back into a plastic container until I buy larger drawers.
But these little wooden drawers (40cm wide by 30x30) however are just the right size for about everything else in the lab... so in just 10 minutes I had found a perfect use for all 3 drawers :

- One is perfect to store all my excess scope probes, 11 of them. I usually only just use a x1/x10 100MHz probe on my Fluke/Philips "Combiscope", my daily driver, so I just keep that probe at all times laying atop the scope, on hand. All other probes I rarely need so I don't need them to clutter the bench or my main / nearby drawers.

- Another drawer is perfect to store all my DMM test leads I just bagged the other day

- Last drawer is perfect to welcome all the coax adapters, BNC and MiniQuick, that I sorted into 3 boxes the other day. This way I keep on hand only a small selection, just what I need to work most of the time.

Really perfect size for most of the stuff in the lab, so I can see myself buying a number of these little drawers soon, as I progressively sort through all my moving boxes to organize all my stuff and bring the lab into operational status so to speak....

So, test leads sorted, power cords sorted, probes sorted, coax adapters sorted, coax cables acquired.... it's going well, lab v2.0 is being shaped a bit more every day 

[xrunner]

Really nice little wooden drawers Vince. Is that a French product or a possible globally available item?

[mnementh]

MuRuPuOrKeViSiVioHaVa

That one is easy ! It's the marking code for a SOT23 5.1V Zener diode ! Everyone knew that I thought ?! 

Nope, is that something you just made up for the sheer hell of it 

Gurrgle thinks it is a geocache of dubious content:

https://www.geocaching.com/geocache/GC4BFK4_murupuorkevisiviohava

mnem
I'd go, but I need to have my scales polished... 

Nope, that geocache description looks fine to me.  Did you not see the electronic components used to make the graphical border around the text?

At a more advanced level, did you see the Winnie-the-pooh reference in that same geocache?     

Honestly, I don't do the geocache thing (the only people I know who do it are even weirder than I am. ) and I was really only referring to the most superficial bit below:

Bogus-koordinaatit vievät luultavasti Martinlaakson voimalaitoksen portille, mutta siellä ei tarvitse käydä.
Bogus coordinates point probably to "power plant", but no need to go there.

Kätkön lähistöltä löydät myös tee-se-itse-kätköilijälle soveltuvan ostospaikankin.
Close to cache you can find also DIY-cacher's shopping place.

Oma kynä mukaan - Bring your own pen.

Yeah, I did... it's too many numbers to be coordinates, and they are all wrong; they mostly start with zero. And end with 01.

Oh wait... are the numbers ASCII codes that map to a Finnish keyboard?

"That's what friends were invented for?"

mnem
maybe I need to smoke some more Smurf pee for it to make sense...

[Vince]

Really nice little wooden drawers Vince. Is that a French product or a possible globally available item?

Oh they are cheap hastily put together badly finished pine wood stuff... but I like wood and they are good enough to me, that I can enjoy them and be happy using them in the lab.

Cost only 12.50 Euros as well, so overall good value for money as far as I am concerned, so bought one to give it at try !

It's probably made in china by some 6 year old slave in a damp basement, like most cheap things... so I would assume you should be able to find something similar near you !

Pine wood is raw of course, because cheap.... so I think I will soon give ti a couple coats of my favorite oil or varnish to protect it, keep it from yellowing, or getting dirty, too fast.

[Kosmic]

For low impedance probes matching is not really an issue because generally "as low a capacitance as possible" is the target and the issue of of HF and LF response balancing is less of a problem as the impedance seen by the probe input is both lower and flatter. To get a flavour of this, let's look at the impedance curve for that TPP1000 10MΩ probe:



You can see that as we get to higher frequencies the input impedance is dominated by the reactance of that 3.9pF input capacitance. The reactance of 3.9pF at 100 MHz is 408Ω, and it is no coincidence that's the region that the probe impedance is in by the time we get to 100MHz. Now, for a probe with 50Ω resistive impedance the 400Ω in parallel to that has a fraction of the effect (about 44Ω total) compared to being in parallel with 10MΩ (calculated as 407Ω for completeness sake) - a reduction of 12% compared to a reduction of 99.996%.

Honestly looking at those graphs just ruins my day.... it means basically you can't probe anything at all really... the impedance drops so quickly, and the phase shit moves so fast before after the narrow flat area, that unless you are looking at a pure sine wave... the signal you see on the scope is hardly what it is actually in the DUT, so making any "measurement" is just a joke 

You get a ballpark idea of what the signal shape might be, OK... other than that, it's just an illusion 

If your signal has any significant amount of energy in harmonics, then be it the rapidly changing impedance/amplitude, or phase shift... the signal displayed on the scope is potentially heavily distorted in all kinds of weird and wonderful ways.

That's mostly why I don't like modern high-speed stuff.... even simply looking at a freaking square wave/digital signal is day dreaming. It will never look like  a square and whatever overshoot or undershoot or dampening or any distortion anywhere on the waveform... you can't tell 100% if it is real, or if it's just the scope and probe setup that creates it for you out of the blue

It is very depressing....

Other reason for not liking high-speed is the price of probes and test gear rises exponentially with the bandwidth.

Sticking to 1GHz max, is about affordable to the hobbyist yet still plenty of stuff to do to learn and have fun with.  High-speed is just a money pit and a recipe for depression.... to me at least ! 

Some people have to do it as a job because all the gizmos are fast digital stuff nowadays.. but as far as fun goes in one's personal lab, I don't see the fun in high-speed digical stuff

Now that I have got that off my chest, let me go back to tying my power cords if I may.

You probably want an active probe then 

Performance will not depress you but the price might 

Yeah they are crazy expensive, but well, in some time, who knows.. once I have all the more basic stuff acquired/ sorted and I start venturing into more fancy and exotic TE... who knows what I might want to acquire !

However, IIRC from a video about active probes a few years back (might well have been that old Tek video, can't remember), even mega expensive active probes aren't magical. Severe limitations and very fragile, no ? ISTR video said they have a very narrow dynamic range, like, ridiculously low, 5 Volts or something, which is ridiculous. Good enough for 3.3 logic stuff I guess, but that's it... unless you add an external atenuator before feeding the signal to the probe, I guess...
So the probe would have an internal mechanism to provide a DC offset voltage so you could shift the input range to where you need it to be for your particular circuit. So say the probe has a 5V dynamic range and you want to measure a signal that evolves between 30 and 35V.. you would tweak the probe to add a 30V offset to it.... and pray that your signal never goes outside that window or IIRC it can get damaged ! 

So.. expensive, yet very limited and cumbersome to use ! 

But well, there is no magic stick so if that's the best science can give to the engineers in the lab... you gotta make do with it eh ! 

Yes you get high input impedance and bandwidth but you loose dynamic range. My LeCroy AP020 got a 10V dynamic range and +-20V offset.

The old active Tek probes normally came with an optional attenuator. For example on the p6202, you could extend the input range from 6V to 60V with the 10X attenuator tip.

[Kosmic]

However, since the sensor is fine, I decided to make an offer on a Boonton Model 4220A digital power meter. The seller listed it for $99 along with "make an offer". I offered $87 and they accepted it. It should arrive today (picture attached). After initial checkout I'll get it all cleaned up and calibrated. The guy I was going to give the Model 42A to was excited and he said he'd gladly pay me back for the 4220A meter.

The saga continues ...

You're getting down into a rabbit hole my friend

To pair your sensor with the 4220 you need a complete AC calibration (from 10dBm to -60dBm). See page 6-9 of the 4220 manual. This could be done with a Boonton 25A or 2520.

If the 4220 has a valid DC calibration (in memory or on paper. normally Boonton shipped the calibration paper inside the meter), I could generate the AC calibration for your sensor if you send it to me (I got a 4300 with integrated calibrator).

[Zucca]

Very sad about the news regarding Keysight and us hobbist/TEA.
It seems they do not give a shit about passionate people with limited budget.

[Vince]

What are you referring to ? I must have missed a post or some piece of news....

But generally companies of all kind don't care about enthusiasts treasuring their old TE, because the company makes money / survives by selling new gear, so they need to please those that actually can buy brand new gear...

[Kosmic]

this https://www.eevblog.com/forum/testgear/keysight-officially-lost-the-plot-dont-buy-if-youre-a-hobbyist/

[Kosmic]

The sad parts is that it used to be possible to get original parts from them. Not anymore I guess 

[xrunner]

You're getting down into a rabbit hole my friend

To pair your sensor with the 4220 you need a complete AC calibration (from 10dBm to -60dBm). See page 6-9 of the 4220 manual. This could be done with a Boonton 25A or 2520.

If the 4220 has a valid DC calibration (in memory or on paper. normally Boonton shipped the calibration paper inside the meter), I could generate the AC calibration for your sensor if you send it to me (I got a 4300 with integrated calibrator).

I don't think it's going to be that bad. I don't have the paperwork inside the meter or any cal data stored. Yet, I'm piping in a 0 dBm reference (50 MHz) from an hp power meter, and using an hp step attenuator to go down from there in 10 dB steps. I just checked the reading with 60 dB attenuation and the 4220A is showing -60.1 dBm. Not bad. All the other steps are within 0.2 dB. I did play with it this afternoon and used an Agilent rf generator to run some tests at higher frequencies and I saw some pretty good measurements. Nothing "wildly" off.

It may not be calibrated to the nth degree but I have to say it's pretty impressive as it is! I'll report more tomorrow.

[mnementh]

Really nice little wooden drawers Vince. Is that a French product or a possible globally available item?

Oh they are cheap hastily put together badly finished pine wood stuff... but I like wood and they are good enough to me, that I can enjoy them and be happy using them in the lab.

Cost only 12.50 Euros as well, so overall good value for money as far as I am concerned, so bought one to give it at try !

It's probably made in china by some 6 year old slave in a damp basement, like most cheap things... so I would assume you should be able to find something similar near you !

Pine wood is raw of course, because cheap.... so I think I will soon give ti a couple coats of my favorite oil or varnish to protect it, keep it from yellowing, or getting dirty, too fast.

reminds me of this one from IKEA:   

https://www.ikea.com/us/en/p/alex-drawer-unit-white-00473546/#content

mnem
moo.

[Vince]

The shape of the opening to open the drawers looks similar indeed... but I guess it's just due to ergonomics 101 rather than a styling  thing as such...

[Neomys Sapiens]

Boonton Model 42A News - The Good, The Bad, and The Ugly

I thought I had it about wrapped up. It had been working perfectly for a week and I even made a lot of progress on an LED light conversion involving a neat translucent mounting plate. Then it developed a problem. The meter would not go all the way back to zero with no rf signal applied and the power meter on (no it is not any sort of problem with the actual meter movement). The meter goes back to zero as it should if the instrument power is turned off. It also would "jump" several dB on the scale while decreasing power in small steps. Every once in a while, this problem goes away for no apparent reason and the instrument will work perfectly again. When the problem is gone, I can verify all the calibration procedure adjustments are spot-on.

One thing I found was that the chopper frequency wasn't 90 Hz but 98 Hz. It is shown as 90 Hz (and sometimes the doubled F of 180 Hz) in many places on the schematic where they show what waveforms are supposed to look like. It is set by a simple unijunction oscillator that has no adjustment.

Which one....I might have it.

Same here..2N6027, 2N2547 and 2N2646 in Stock

[Vince]

this https://www.eevblog.com/forum/testgear/keysight-officially-lost-the-plot-dont-buy-if-youre-a-hobbyist/

Thanks for the link. Wow 10 pages worth of messages... looked a the first few and last few.

Crap indeed, but looks like people think it's not really a Keysight specific thing and will probably spread to every manufacturer eventually, due to US and EU regulations.

So soon we will have other similar threads about other brands...

However where I see Keysight has no excuse, is no providing support to the OP in that thread, even though he did buy his scope brand new ! And not an old one at that. It's not like he was trying to get support and parts for 40 year old spectrum analyzer or something...

[mnementh]

   Phew, done with the power cords ! 

Nice little collection of ball-busters you've got there, V! 

mnem
so glad I only have a few of those that made their way to me by accident...

[xrunner]

Which one....I might have it.

Same here..2N6027, 2N2547 and 2N2646 in Stock

It's a 2N4871, but I don't need it because I substituted a pot for the resistive element controlling the freq. I do want to acknowledge your kind offer though I appreciate it a lot.