Author Topic: HP/Keysight 66311B magnetic field leakage  (Read 4987 times)

0 Members and 1 Guest are viewing this topic.

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
HP/Keysight 66311B magnetic field leakage
« on: November 16, 2023, 03:10:50 pm »
Hello,

i've recently bought a used HP 66311B Mobile Communications DC Source from around 2003, which is working pretty well.
But when i power it on, the case of the supply is making a relatively loud hum and when i place it near my crt oscilloscope the whole picture is shaking wildly. The Souce needs to be at least 40 cm away from the scope to not distort the screen.
Thats why i suspect excessive magnetix flux leakage from the transformer.

Has anyone else the same problem?
Any tips on how to improve the situation?

It's very annoying, because i want to use the scope and the supply on the same eqipment stack.
I already removed all the secondary connections from the transformer, but its still happening.


Greetings from germany
 

Offline beatman

  • Regular Contributor
  • *
  • Posts: 65
  • Country: gr
Re: HP/Keysight 66311B magnetic field leakage
« Reply #1 on: November 16, 2023, 03:53:18 pm »
one solution is to separate the trans from enclosure to another metal box under your work desk and do the connections with thick cables to the main unit.one other try is to cover carefully the transformer with copper or iron foil to isolate the flux that escape.i am not sure that works.may you get some more options from other guys here.
 

Offline BlownUpCapacitor

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: us
  • I kinda suck at the art of electronics tbh
Re: HP/Keysight 66311B magnetic field leakage
« Reply #2 on: November 16, 2023, 06:51:02 pm »
What beatman said is correct. Adding an EMI shield will definitely help.

Use copper tape to make a strap around the switching transformer as shown in the image. Then ground it.

If that doesn't work you can try using mu-metal shielding to stop EMI. https://en.wikipedia.org/wiki/Mu-metal

Though I'm not too sure of its effectiveness at high frequencies... maybe someone can answer that for me.

Edit: Shouldn't the analog scope have mu-metal shielding already though?
« Last Edit: November 16, 2023, 06:53:14 pm by BlownUpCapacitor »
Hehe, spooked my friends with an exploding electrolytic capacitor the other day 😁.
 

Offline AVGresponding

  • Super Contributor
  • ***
  • Posts: 4931
  • Country: england
  • Exploring Rabbit Holes Since The 1970s
Re: HP/Keysight 66311B magnetic field leakage
« Reply #3 on: November 16, 2023, 06:53:50 pm »
WTH! This is definitely not normal, mine does not do this. The only noise it makes is the fan, which only gets annoying when you load it up a bit.
nuqDaq yuch Dapol?
Addiction count: Agilent-AVO-BlackStar-Brymen-Chauvin Arnoux-Fluke-GenRad-Hameg-HP-Keithley-IsoTech-Mastech-Megger-Metrix-Micronta-Racal-RFL-Siglent-Solartron-Tektronix-Thurlby-Time Electronics-TTi-UniT
 

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #4 on: November 20, 2023, 07:39:17 am »
What beatman said is correct. Adding an EMI shield will definitely help.

Use copper tape to make a strap around the switching transformer as shown in the image. Then ground it.

If that doesn't work you can try using mu-metal shielding to stop EMI. https://en.wikipedia.org/wiki/Mu-metal

Though I'm not too sure of its effectiveness at high frequencies... maybe someone can answer that for me.

Edit: Shouldn't the analog scope have mu-metal shielding already though?

I will try it, thanks.
I found some mu metal foil on ebay.

Its a Tektronix TDS460A, so not even a bad scope. But still...

WTH! This is definitely not normal, mine does not do this. The only noise it makes is the fan, which only gets annoying when you load it up a bit.

I thought so, but what can be the issue, when the supply is working normally, the voltages are matching and the transformer is leaking, even under no load.
It must either be some kind of production issue, or the core of the transformer is broken internally and the flux needs to jump an air-gap or whatever.
 

Offline AVGresponding

  • Super Contributor
  • ***
  • Posts: 4931
  • Country: england
  • Exploring Rabbit Holes Since The 1970s
Re: HP/Keysight 66311B magnetic field leakage
« Reply #5 on: November 20, 2023, 10:24:05 am »
What beatman said is correct. Adding an EMI shield will definitely help.

Use copper tape to make a strap around the switching transformer as shown in the image. Then ground it.

If that doesn't work you can try using mu-metal shielding to stop EMI. https://en.wikipedia.org/wiki/Mu-metal

Though I'm not too sure of its effectiveness at high frequencies... maybe someone can answer that for me.

Edit: Shouldn't the analog scope have mu-metal shielding already though?

I will try it, thanks.
I found some mu metal foil on ebay.

Its a Tektronix TDS460A, so not even a bad scope. But still...

WTH! This is definitely not normal, mine does not do this. The only noise it makes is the fan, which only gets annoying when you load it up a bit.

I thought so, but what can be the issue, when the supply is working normally, the voltages are matching and the transformer is leaking, even under no load.
It must either be some kind of production issue, or the core of the transformer is broken internally and the flux needs to jump an air-gap or whatever.

Maybe the shield is disconnected, removed, or something odd like that?

I tried to recreate the effect you're seeing; easier said than done without moving quite a bit of heavy stuff. As it happens my TDS420A tube is over 70cm away from the transformer in the 66311B, so the most practical way I could think of was to bring a small tube very close to the psu.
I have a Thandar SC110 and I brought it right up to the side of the 66311B, switched it on and off, turned the output on and off, the trace on the scope didn't move at all.
nuqDaq yuch Dapol?
Addiction count: Agilent-AVO-BlackStar-Brymen-Chauvin Arnoux-Fluke-GenRad-Hameg-HP-Keithley-IsoTech-Mastech-Megger-Metrix-Micronta-Racal-RFL-Siglent-Solartron-Tektronix-Thurlby-Time Electronics-TTi-UniT
 

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #6 on: November 21, 2023, 09:43:23 am »
So it must be the transformer.
I removed all secondary connections from the transformer and let it run with just the primary connected.
The power draw is around 3 watts in this condition and 16 watts with all the secondaries connected....
The effect is still the same.
I tried to place my Tek 2230 beside it as a sanity check and the screen also wiggles.
In the attached picture you can the the wiggle in the bottom half of the crt, where the lines are 2-3 mm apart.

I tried to ground the shielding and the core with a cable to chassis, but no change.

The transformer is an 9100-5276, which is the correct number for this model and the voltages are all within resonable margins.

Maybe mu metal would help, but somehow the transformer must be bad.

 

Offline BlownUpCapacitor

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: us
  • I kinda suck at the art of electronics tbh
Re: HP/Keysight 66311B magnetic field leakage
« Reply #7 on: November 21, 2023, 05:10:21 pm »
The transformer is a mains transformer? I thought it was switching!

Though... a linear power supply would make much much more sense for a Mobile Communications DC powersupply to reduce the EMI.

Not too sure why I thought it was switching....

Must have been playing around with the cheap lab bench power supplies for too long.
Hehe, spooked my friends with an exploding electrolytic capacitor the other day 😁.
 

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #8 on: November 23, 2023, 09:58:18 am »
So with this amount of leakage and hum, the unit has no use to me....

I've looked into the service manual and it seems that i have to have 4x 24V rails and one 12V rail to power the unit.
On a scale from 1 to 10, how dumb would be a fully reversable mod using these chinese isolated power supplys?....



All the connectors on the transformer are socketed, so i can use spade connectors to adapt them, and a 3d printed mounting bracket to interface the original transformer mounting.

In germany it would be impossible to get the original transformer as a replacement part for less then what i paid for that power supply.
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 11336
  • Country: us
  • $
Re: HP/Keysight 66311B magnetic field leakage
« Reply #9 on: November 23, 2023, 10:11:31 am »
thats not normal. are you sure you have a ground? maybe your earth ground is really bad or something. like grounded through rusted conduit.


not really sure what can go wrong with a transformer like that to make it leak so much field either. tighten the bolts?


have you tried moving both instruments to another outlet
« Last Edit: November 23, 2023, 10:17:12 am by coppercone2 »
 

Offline srb1954

  • Super Contributor
  • ***
  • Posts: 1124
  • Country: nz
  • Retired Electronics Design Engineer
Re: HP/Keysight 66311B magnetic field leakage
« Reply #10 on: November 23, 2023, 10:27:18 am »
So it must be the transformer.
I removed all secondary connections from the transformer and let it run with just the primary connected.
The power draw is around 3 watts in this condition and 16 watts with all the secondaries connected....
The effect is still the same.
Have you got a current probe of current transformer to measure the primary current?

An excessively spiky current might indicate that the transformer core is saturating. The permability of the iron would then reduce and there would be more stray magnetic field field outside of the core.

Another option is to use a variac to reduce the primary voltage a little and see if that reduces the stray field. A small drop in the primary voltage would significantly reduce any core saturation effects if present.
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: HP/Keysight 66311B magnetic field leakage
« Reply #11 on: November 23, 2023, 12:08:11 pm »
I think I have found the problem. There is a metal (steel / iron) shield missing around the transformer. Mine looks like this:

« Last Edit: November 23, 2023, 12:12:35 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline electronomicon

  • Contributor
  • Posts: 34
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #12 on: November 23, 2023, 12:30:59 pm »
I’ve got the same problem on different devices. For me, it only shows up in the combination of two similar HP/Agilent devices (HP 33120A and Agilent 34401A, either one of them) and one scope (HP 54645D). Not with my old Philips scope and neither any other than those two devices with the 54645D. It’s definitely very annoying.

I also did some investigation, like the thread starter, and it comes back to the mains transformer, too. I disconnected everything inside the 33120A, but as soon as the mains transformer has power and is close to the scope, the CRT starts wiggling around.
The transformer is open in both devices (see attached image) … should there be a shield around it that I’m missing?

I might try to wrap the transformer, to see if that helps. What thickness of copper foil would be advisable to use?
« Last Edit: November 23, 2023, 12:41:06 pm by electronomicon »
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28429
  • Country: nl
    • NCT Developments
Re: HP/Keysight 66311B magnetic field leakage
« Reply #13 on: November 23, 2023, 01:25:43 pm »
Copper foil won't do you much good as the problem is the magnetic field. You need to use a material which is good at transporting magnetic fields. Like iron or mu-metal.

In this thread I experimented with adding an extra shield around a transformer with good results:
https://www.eevblog.com/forum/testgear/korad-kel2010-multicomp-mp710771-review/msg5039434/#msg5039434

In the end I also moved the transformer away from sensitive circuits but the shield by itself improved the situation significantly.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #14 on: November 23, 2023, 02:08:55 pm »
I think I have found the problem. There is a metal (steel / iron) shield missing around the transformer. Mine looks like this:


Thanks for the suggestion, but the shield is normally there.
I just removed it to read the serial number and take a good picture of the transformer.

I’ve got the same problem on different devices. For me, it only shows up in the combination of two similar HP/Agilent devices (HP 33120A and Agilent 34401A, either one of them) and one scope (HP 54645D). Not with my old Philips scope and neither any other than those two devices with the 54645D. It’s definitely very annoying.

I also did some investigation, like the thread starter, and it comes back to the mains transformer, too. I disconnected everything inside the 33120A, but as soon as the mains transformer has power and is close to the scope, the CRT starts wiggling around.
The transformer is open in both devices (see attached image) … should there be a shield around it that I’m missing?

I might try to wrap the transformer, to see if that helps. What thickness of copper foil would be advisable to use?

When scrolling through the service manual it seems they used the same transformer in all versions HP, Agilent and Keysight, it seems weired, that no one noticed this in the more or less 30 years of manufacturing time.

Maybe mu-metal is the best option, but 30% reduction would not be sufficient i guess.
I also have an 34401A, i will try it near the scope, when i'm back at the bench.

 

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: HP/Keysight 66311B magnetic field leakage
« Reply #15 on: November 23, 2023, 11:03:11 pm »
Make sure you have the voltage set correctly, e.g. 240 for europe rather than 220 (go higher rather than lower). Flux leakage often gets a LOT worse the closer it is to saturation.
Mine does make enough humming noise on the ~245-250V they deliver to me that I'm not surprised you're seeing issues near CRTs.
 
The following users thanked this post: nctnico, electronomicon

Offline electronomicon

  • Contributor
  • Posts: 34
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #16 on: November 24, 2023, 12:04:26 am »
Make sure you have the voltage set correctly, e.g. 240 for europe rather than 220 (go higher rather than lower). Flux leakage often gets a LOT worse the closer it is to saturation.
Mine does make enough humming noise on the ~245-250V they deliver to me that I'm not surprised you're seeing issues near CRTs.

Thanks for the suggestion! I just set my 33120A (the upper device in the stackup) from 220 V to 240 V and its effect on the CRT is now barely noticable. I hope with some additional shielding on the 34401A’s transformer it will be little enough to easily ignore it.
 

Offline electronomicon

  • Contributor
  • Posts: 34
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #17 on: November 29, 2023, 11:05:26 pm »
So I ordered some mu-metal sheet and applied it today. For each of my two noisy devices one layer directly over the lower side of the transformer and one on the housing beneath (see attached images).
It definitely helps. With just the device closer to the scope turned on it’s just about noticeable, with only the farther one turned on not at all (when I previously had the shielding only applied to the closer one, the farther one made more noise than the closer one). Interestingly, when I turn both on, the noise is most noticeable (the farther one here seems to increase the field, despite not having an effect when run on its own). But even then it’s still acceptable and probably less than only the farther one before.

Maybe it would help to also add a piece of mu-metal to the top of the scope housing? But honestly, with the cost of this stuff I guess I’m happy with it right now (I was lucky and got a piece from an ebay auction comparatively cheap).

Edit: But if you attempt to do something like this, too, be careful with the edges. I cut my sheet with a paper cutter, that worked out great, but it left an edge sharp enough to cut myself in the finger |O.
« Last Edit: November 29, 2023, 11:12:28 pm by electronomicon »
 
The following users thanked this post: Gulftown

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #18 on: December 01, 2023, 11:12:25 am »
Okay that sounds pretty good.
I just ordered some mu metal as well.

I was thinking about reducing the input voltage with 2 20V 600W zener diodes on either side of the transformer inputs. That should limit the input voltage, but introduce some side effects due to the dead time and non sinusoidal waveform in the middle.
Do you think this would work?
Or would it do more harm then good?

 
 

Offline GulftownTopic starter

  • Contributor
  • Posts: 42
  • Country: de
Re: HP/Keysight 66311B magnetic field leakage
« Reply #19 on: December 01, 2023, 11:27:31 am »
Okay I found a solution I think.
I will use a 230V/24V transformer and use it as an bucking transformer to reduce the input voltage on the primary of the original transformer.
That should work without introducing side effects.
I will keep you updated.
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 11336
  • Country: us
  • $
Re: HP/Keysight 66311B magnetic field leakage
« Reply #20 on: December 02, 2023, 06:41:09 am »
use sand paper blocks to deburr sheet metal edges. IMO don't bother even getting near sheet metal if you don't have sanding blocks of the right size ready, its asking for a bad day

this thread makes me want to get a sheet metal mini deburring tool
https://www.aircraftspruce.com/catalog/topages/dbledeb.php

and I also want to try to etch mu-metal sheet now.
« Last Edit: December 02, 2023, 06:45:06 am by coppercone2 »
 

Offline AVGresponding

  • Super Contributor
  • ***
  • Posts: 4931
  • Country: england
  • Exploring Rabbit Holes Since The 1970s
Re: HP/Keysight 66311B magnetic field leakage
« Reply #21 on: December 02, 2023, 08:39:45 am »
I prefer to use a file. Especially if I'm on a job fabricating a lot of metal trunking, sandpaper/emery paper wouldn't be durable enough.
nuqDaq yuch Dapol?
Addiction count: Agilent-AVO-BlackStar-Brymen-Chauvin Arnoux-Fluke-GenRad-Hameg-HP-Keithley-IsoTech-Mastech-Megger-Metrix-Micronta-Racal-RFL-Siglent-Solartron-Tektronix-Thurlby-Time Electronics-TTi-UniT
 

Offline coppercone2

  • Super Contributor
  • ***
  • Posts: 11336
  • Country: us
  • $
Re: HP/Keysight 66311B magnetic field leakage
« Reply #22 on: December 02, 2023, 09:02:16 am »
it is easy to mangle with a file if you dont make a good cut but the sand paper solution is quite dodgy like you say.

stones might also be an option (machinist stone)

Diamond file is a good option too, but slow. otherwise get a  very fine file (typically small). or sanding sticks but they will get torn up too
 
The following users thanked this post: electronomicon

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: HP/Keysight 66311B magnetic field leakage
« Reply #23 on: December 02, 2023, 01:13:45 pm »
Okay I found a solution I think.
I will use a 230V/24V transformer and use it as an bucking transformer to reduce the input voltage on the primary of the original transformer.
That should work without introducing side effects.
I will keep you updated.
Bucking transformer is indeed the right way to go if you need to drop mains voltage a little. If you do it in the configuration shown in Fig. 4 here then you also avoid saturation issues on the bucking transformer: https://sound-au.com/articles/buck-xfmr.htm

I would test by dropping the voltage a bit with a variac first though to be sure it's actually going to solve the issue before spending the time and money wiring up the bucking transformer.
 
The following users thanked this post: Gulftown

Offline Hydron

  • Super Contributor
  • ***
  • Posts: 1085
  • Country: gb
Re: HP/Keysight 66311B magnetic field leakage
« Reply #24 on: December 02, 2023, 01:18:15 pm »
Make sure you have the voltage set correctly, e.g. 240 for europe rather than 220 (go higher rather than lower). Flux leakage often gets a LOT worse the closer it is to saturation.
Mine does make enough humming noise on the ~245-250V they deliver to me that I'm not surprised you're seeing issues near CRTs.

Thanks for the suggestion! I just set my 33120A (the upper device in the stackup) from 220 V to 240 V and its effect on the CRT is now barely noticable. I hope with some additional shielding on the 34401A’s transformer it will be little enough to easily ignore it.
Does the 34401A not also have a 240V line voltage setting? Or was that not enough to fix the issue? Glad the suggestion helped with the other unit.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf