Author Topic: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review  (Read 13255 times)

0 Members and 1 Guest are viewing this topic.

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
A possible future project made me want to look for a device which can analyse inductors over a frequency range. At first I was looking at LCR meters but the more flexible ones are expensive. Then I thought I might be able to do the same with a device called a network analyser so I started looking for those. Last week this Anritsu MS4630B caught my eye. The 'calibration void' stickers looked intact and it was listed as 'doesn't power up'.


After some haggling I got it for $400 + shipping and it arrived today. Time for some testing and crap(!)... it does power up (fan running) but it doesn't display anything. That is bad because it means there could be a really serious problem instead of the simple SMPS fix I was hoping for. So lets open it up and see what is inside:

From left to right: power supply, processing unit (with cover removed) and HF section. First I checked the voltage on the batteries but these where fine. Next thing on the fault finding list: check all the connections. And BINGO! The connector with the cable to the front panel isn't properly seated so I pushed it in and tried to turn it on again:

Woohooo! YES! It powers up and all self tests passed! Great!

Now let's play a little bit with it! A network analyser is very closely related to a spectrum analyser; it basically is a spectrum analyser but it has a more precise tracking generator and 2 spectrum analysis inputs (in the case of the MS4630B switchable between 50 Ohm and 1M Ohm impedance). The MS4630B I have also has the 70dB (10dB steps) output attenuator option installed which makes measuring (for example) amplifiers a whole lot easier. You may ask yourself: what can a network analyser like this do? Well it can do many things:
- Be used as a spectrum analyser and since the MS4630B goes down to 10Hz (3Hz RBW) it is also suitable for audio
- Determine the transfer function (magnitude and phase) of passive and active circuits
- Measure equivalent circuits of inductors and capacitors (and networks) with the use of a directional bridge (already ordered).

Allowing the input impedance to be set to 1M Ohm makes it very easy to hook it up to about any circuit and take a measurement.

First I measured an elliptic high pass LC filter I made to filter 50Hz from a large signal:


Then I went for the classic RC high pass filter (1k Ohm / 10nf) and also plotted the phase:


There still is one problem I'll need to solve though: how to get a screendump from it. The disk drive isn't working (media error) but even if it worked I immediately hit the next problem: my PC no longer has a disk drive and the two USB disk drives I have tried so far where a total dissaster. It can also use a new power button. Hopefully I can get away with a standard cap.

Still I think this is a nice catch and may turn out to be a better buy than an LCR meter for my purpose (although the LCR meter would be more precise).
« Last Edit: July 11, 2016, 08:55:53 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline PA0PBZ

  • Super Contributor
  • ***
  • Posts: 5121
  • Country: nl
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #1 on: July 11, 2016, 08:57:12 pm »
Nice score Nico! If you can get the disk working (shouldn't be too difficult to replace it with a working one) you could always get an USB 3.5" floppy drive for your PC.
Keyboard error: Press F1 to continue.
 
The following users thanked this post: Samba

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11228
  • Country: us
    • Personal site
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #2 on: July 11, 2016, 09:08:53 pm »
There still is one problem I'll need to solve though: how to get a screendump from it. The disk drive isn't working (media error) but even if it worked I immediately hit the next problem: my PC no longer has a disk drive and the two USB disk drives I have tried so far where a total dissaster.
There are USB / SD Card adapters that fit into standard disk drive for people running old DOS games. I'm sure eBay will help you here.
Alex
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #3 on: July 11, 2016, 09:15:56 pm »
There still is one problem I'll need to solve though: how to get a screendump from it. The disk drive isn't working (media error) but even if it worked I immediately hit the next problem: my PC no longer has a disk drive and the two USB disk drives I have tried so far where a total dissaster.
There are USB / SD Card adapters that fit into standard disk drive for people running old DOS games. I'm sure eBay will help you here.
I have tried one of those in an oscilloscope but it was extremely tedious to use. Also the USB disk drives I bought for my PC where total crap. The first was DOA and the second one from a somewhat reputable brand (the most expensive I could find at my IT parts supplier) died the 5th or 6th time I used it. I don't mind spending the money but the quality products don't seem to be there.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online ataradov

  • Super Contributor
  • ***
  • Posts: 11228
  • Country: us
    • Personal site
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #4 on: July 11, 2016, 09:22:39 pm »
I have tried one of those in an oscilloscope but it was extremely tedious to use. Also the USB disk drives I bought for my PC where total crap. The first was DOA and the second one from a somewhat reputable brand (the most expensive I could find at my IT parts supplier) died the 5th or 6th time I used it. I don't mind spending the money but the quality products don't seem to be there.
Well, I don't know, I have not tried them myself, but people who run DOS games on old PCs seem to be fine with the way they work.
Alex
 

Online TheSteve

  • Supporter
  • ****
  • Posts: 3743
  • Country: ca
  • Living the Dream
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #5 on: July 11, 2016, 09:41:51 pm »
We replaced the floppy drive in our DOS 3.3 based Juki 570L pick and place machine with one of the USB floppy emulators and after finding the right jumper settings it worked great. I would say to get one of the basic models that just emulates a single floppy image on your flash drive instead the ones that allow up to 99 etc as they need a program or driver to read the virtual floppies on the flash drive using your PC - that isn't worth the headache.
VE7FM
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #6 on: July 11, 2016, 10:46:30 pm »
Now I'm tempted to locate that USB floppy emulator and try again. IIRC it had something to do with the unlabled buttons. One is for copying the internal memory to USB and the other vice versa. Press the wrong button and all your screendumps are gone  :rant:

Edit: tried the emulator but no go. It could be the floppy controller is bad though.

Edit2: I will not be defeated! I took the diskdrive out to look for the manufacturer and part number. It turns out it is a TEAC FD-05HG slimline diskdrive. So fire up Ebay and from some listings I learned these drives are also used in several Dell machines. Hmm... I also have several Dell machines so a dive into the stack with retired Dell machines quickly got me a replacement disk drive! I tried it but still no-go. Drats! After some more fooling around in the hardcopy menus I found a menu to format a disk (I'm using formatted disks!). Lets try that... OK formatting works... so far so good... lets try and make a screendump again.. and wow... now it works!
« Last Edit: July 12, 2016, 01:07:20 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline diarl

  • Newbie
  • Posts: 1
  • Country: us
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #7 on: July 19, 2016, 02:56:16 pm »
Hi nctnico, congratulations on the new VNA! You don't need a disk drive for it. Check the GPIB port located on the back of the machine. You can always connect it to a computer to do all the communication and control. Another thing is, you need to calibrate the machine before using it. Check the SOLD(short/open/load/through) calibration
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #8 on: July 19, 2016, 04:29:42 pm »
I have a GPIB interface but how about the software to control the MS4630B? I always use screendumps in reports and/or I write the measurement results down because I have not had much luck with software to control instruments remotely. Most of it is just cumbersome to setup and by the time it is setup I have already copied 10 screendumps into a report.
« Last Edit: July 19, 2016, 05:15:02 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online edavid

  • Super Contributor
  • ***
  • Posts: 3381
  • Country: us
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #9 on: July 19, 2016, 05:43:49 pm »
Also the USB disk drives I bought for my PC where total crap. The first was DOA and the second one from a somewhat reputable brand (the most expensive I could find at my IT parts supplier) died the 5th or 6th time I used it. I don't mind spending the money but the quality products don't seem to be there.

Look for a Dell USB floppy drive.  The one I have is marked Teac FD-05PUB on the bottom.  It works fine.

They are very cheap in the US:

https://www.amazon.com/Dell-1-44MB-External-Floppy-Drive/dp/B000I2XGCY

https://www.amazon.com/1-44MB-External-Floppy-Drive-Black/dp/B000M3GODW
« Last Edit: July 19, 2016, 05:46:13 pm by edavid »
 

Offline PA0PBZ

  • Super Contributor
  • ***
  • Posts: 5121
  • Country: nl
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #10 on: July 19, 2016, 07:10:03 pm »
Look for a Dell USB floppy drive.  The one I have is marked Teac FD-05PUB on the bottom.  It works fine.

Yes, I have the same (came with a server a long time ago) and it works fine.
Keyboard error: Press F1 to continue.
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #11 on: July 19, 2016, 07:40:13 pm »
OK, I'll go on the hunt for one of those Dell USB diskdrives! Thanks!
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Online voltsandjolts

  • Supporter
  • ****
  • Posts: 2277
  • Country: gb
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #12 on: July 19, 2016, 08:18:55 pm »
Nice piece of kit and very lucky to have such an easy repair!

I can supply you with a brand new TEAC Floppy Drive FD-05HF for a modest fee, if thats any good to you.
They are a slim drive often used in test equip...and sewing machines!
« Last Edit: July 19, 2016, 08:28:43 pm by voltsandjolts »
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #13 on: July 19, 2016, 08:39:49 pm »
Thanks for the offer but I already found a (working) replacement floppy drive in my stack of retired Dell systems.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline R_G_B_

  • Frequent Contributor
  • **
  • Posts: 399
  • Country: gb
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #14 on: July 21, 2016, 09:39:27 pm »
I have the older version of this network analyzer. The frequency response is not as wide 10Hz.....30Mhz.

Also I can not seem to get it to work at low frequencies 50Hz for some strange reason.

Works well other than this.

http://www.ebay.com/itm/Anritsu-MS3401A-10Hz-30MHz-Network-Analyzer-/141821547885


R_G_B
R_G_B
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #15 on: July 21, 2016, 10:23:17 pm »
What kind of circuit did you try to analyse? Perhaps the signals got too small to resolve. I used to have an higher end HP LCR meter (big boat anchor) but measuring resistors in the milli-Ohm range just wasn't possible with it.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline R_G_B_

  • Frequent Contributor
  • **
  • Posts: 399
  • Country: gb
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #16 on: July 22, 2016, 06:37:35 pm »
It was just a RC low pass filter.
R_G_B
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #17 on: September 04, 2016, 01:40:36 pm »
Yesterday I could no longer stand the missing power button. Unfortunately I have been unable to find a replacement so I glued some PVC sheet together and milled it into the right shape:


Making the square hole at the back was a bit iffy but I used a 1.5mm mill bit to make a hole square-ish


Much better!
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline R_G_B_

  • Frequent Contributor
  • **
  • Posts: 399
  • Country: gb
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #18 on: September 08, 2016, 09:24:36 pm »
I have the predecessor to this network analyzer.

I managed to sort the problem with it which is the sweep speed. If its set to a fast sweep speed and the filter has a very low frequency band it would not plot. So you have to slow down the sweep speed in order for it to plot the filter response.

R_G_B
 

Online PartialDischarge

  • Super Contributor
  • ***
  • Posts: 1611
  • Country: 00
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #19 on: November 02, 2016, 07:22:20 pm »
Nice equipment.
Is there anything similar from other brands? I'm looking for something that can measure phase and freq response of a DUT from nearly dc to some 10-100MHz.
 

Online edavid

  • Super Contributor
  • ***
  • Posts: 3381
  • Country: us
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #20 on: November 02, 2016, 07:33:10 pm »
Nice equipment.
Is there anything similar from other brands? I'm looking for something that can measure phase and freq response of a DUT from nearly dc to some 10-100MHz.

HP 3577A or HP 3589A network analyzer

HP 3575A gain/phase meter
 

Online PartialDischarge

  • Super Contributor
  • ***
  • Posts: 1611
  • Country: 00
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #21 on: November 02, 2016, 07:55:08 pm »
You just made my day edavid. What a beauty that 3575A, old style HP, no software updates or EMI crap problems. I may get one just for the fun of it
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1388
  • Country: de
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #22 on: May 09, 2017, 05:56:58 pm »
I've got an inquiry to all the MS4630B owners/users:

We've recently acquired such a machine at work (basically for playing around a little...) but stumbled quickly into some strange behavior:

It actually seems to be a software bug which may be related to an early software revision on our analyzer. What we basically found was a strange behavior of our DUT's delay which should be fairly constant over the frequency. We narrowed the problem down to a comparison of two BNC cables of different lenght (a 0.5m one from out B to the R(eference) input and a 2m one from the second out B to the T(est)A input). If we set up the machine to measure magnitude and delay while the frequency axis is linear, everything's fine and we see the delay difference as a flat line, offset by the delay of the electrical length difference of the two cables.

As soon as we change the frequency axis to logarithmic, the delay is no longer a flat line but keeps on increasing to the edge of the window. Please see attached photos for the setup and the findings (sorry for the poor quality, bad lighting conditions, cell phone cam and a somewhat tired display backlight...).

What I would like to know is if other machines behave the same way or if this bug had been fixed in a later software revision. Does anybody know if there's a firmware update available and if the firmware could be user-updated at all? Does anyone know of a service manual for these analyzers?

Thanks and all the best,
Thomas
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #23 on: May 09, 2017, 06:57:04 pm »
I did some tests on mine (which seems to have slightly newer firmware). I think you can't measure delay like this. You have to remember that everything this machine measures is based on phase and amplitude. At 10kHz the phase difference caused by a 2m long piece of coax is extremely small.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline 1design

  • Regular Contributor
  • *
  • Posts: 162
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #24 on: May 09, 2017, 07:32:45 pm »
I would guess what you see is the difference in behaviour of the two cables. Since you did not calibrate the unit and it compares the two inputs, it shows the difference in behaviour of the two cables, which in this case is frequency dependent.

Also, when you change the scale to logarithmic you zoom into the upper decade, magnifying the portion of the spectrum that has the transient behaviour typical of low pass filters(cables). In the linear scale that is probably  just a few pixels on the edge of the screen, try to re-scale it in linear mode.
 

Online TurboTom

  • Super Contributor
  • ***
  • Posts: 1388
  • Country: de
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #25 on: May 09, 2017, 08:31:39 pm »
Initially, we (my collegue -- the "real" EE -- and me, the "crazy" physicist...) also questioned our setup and what errors we might have done when we found this discrepancy. Truth is, the BNC cables we used are good enough at the frequency range of that device that we shouldn't see any deviations of the delay. I checked the cables with my HP 8753C and they perform flawlessly up to > 1GHz delay-wise, above that some tiny wiggles can be observed, but still no significant deviation from the general phase delay, up to the 8753C's limit of 3GHz. Btw, the 8753 doesn't show the "funny" incline of the delay plot in logarithmic frequency mode.

Of course, the delay will be introduced by the length difference of the two cables, that was the intended effect. I also agree that the 4630B (like probably most VNAs) primarily evaluate phase (by measuring the balance between the I and Q signals) but can easily calculate phase delay by dividing continuous phase by (360°*frequency). If the DUT (cable's dielectric) hasn't got any significant dispersion in the observed frequency range and there aren't any resonances otherwise, this figure should be fairly constant as it actually is, provided the linear frequency scale is selected (see attached photo).

For comparison, I attached the phase measurement of our two cables as well, both with linear and with logarithmic frequency scale. These figures are both correct. To me, it actually seems that Anritsu erroneously used the frequency reading from the linear scale to calculate the delay in the logarithmic display, which would also explain that right at the end of the logarithmic delay plot, the reading drops back to the approximately correct range.

Nico, what's the firmware date and revision of your 4630B if i may ask?

Thanks and all the best,
Thomas



Edit: Added the delay plot with logarithmic time scale for better comparison.
« Last Edit: May 09, 2017, 08:34:37 pm by TurboTom »
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #26 on: May 09, 2017, 09:36:15 pm »
I'm getting similar weird readings on mine. Also my MS4630B only wants to measure the delay up to 100MHz  :-//
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Omicron

  • Regular Contributor
  • *
  • Posts: 142
  • Country: be
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #27 on: August 04, 2017, 07:45:54 pm »
I have one of these as well. However I think mine is faulty. The traces are really noisy in the region from 10Hz to 50KHz or so, jumping up and down by about 10dB. Calibration does not flatten out the trace in this band as it does for frequencies above 50KHz and the calibrated trace is just as noisy. I'm assuming there must be a fault in the LF part of the analyser. Unless I'm missing something obvious? I've only spent about an hour playing with the unit.

Has anyone stumbled upon a service manual for these things?

[Edit] Ah! My bad! I need to select a longer sweep speed if I want to see the lower end of the spectrum. Obvious in retrospect! :-)
« Last Edit: August 04, 2017, 08:43:31 pm by Omicron »
 

Offline R_G_B_

  • Frequent Contributor
  • **
  • Posts: 399
  • Country: gb
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #28 on: August 04, 2017, 11:01:51 pm »
I have the predecessor to this network analyzer.

And it shows something similar in the low frequency band
Not sure if it's could be the resolution bandwidth settings Or sweep time settings.

R_G_B_
R_G_B
 

Offline Omicron

  • Regular Contributor
  • *
  • Posts: 142
  • Country: be
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #29 on: August 04, 2017, 11:33:39 pm »
If I turn the sweep speed down the trace clears up. When I set it to 100 seconds for example then I get a pretty clean trace all the way down to 10Hz. It makes sense if you think about it, the sweep needs to be slow enough for it to be able to capture a few complete sine wave cycles at each measurement point. So the lower you go the more time it needs.

I'm still learning how to use this device, but I'm pretty sure now mine is working as intended.
 

Offline mrf245

  • Regular Contributor
  • *
  • Posts: 79
  • Country: cn
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #30 on: April 23, 2018, 12:33:50 am »
I bought some MS4630B with TRB (secondary receiver port), some of them are with issues such as LCD fault and panel broken. Now trying to repair it.
BH7JUO Mark
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #31 on: April 23, 2018, 08:28:36 pm »
Replacing the LCD panel shouldn't be hard. It is a standard VGA panel.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline nctnicoTopic starter

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: Anritsu MS4630B 10Hz-300MHz network analyser teardown/repair/review
« Reply #32 on: August 20, 2022, 12:27:16 pm »
Ressurecting my old thread...
A couple of years ago I bought a Prologix ethernet to USB adapter to get screendumps from a Tektronix frequency counter. Since then I have created several other Python scripts to get screendumps from other equipment as well. The latest addition is the MS4630B.

Or sort off... it turns out it can't really do a screendump over GPIB. It can only print. I have looked at printer emulation in software or hardware but that is either gone, expensive or requires different hardware. So I ended up creating a Python script that reads the data, plots a graph and saves it to disk. The attached script works OK for x-y plots both linear and logarithmic. It took a bit of effort to make Matplotlib use sensible numbers for the x axis.  Polar and smitch charts aren't implemented.

Some examples:





Edit: added updated script which also shows the time + date and allows for an (optional) title using a -t option on the command line.
« Last Edit: August 21, 2022, 04:52:09 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: jjoonathan


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf