Author Topic: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked  (Read 3045 times)

0 Members and 1 Guest are viewing this topic.

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Hello. With my ESG 4433B I have a problem with the digital modulations.
When activating the dual ARB (option UND) I see the <error 518 unlocked> and UNLEVEL.

It does not matter if the setting for the "ARB Reference" key is set to EXT or INT.
Externally, no source is connected.

The ARB seems to work. Digital modulations are generated.
These sound the same regardless of ARB Reference selection on EXT or INT.

After about 2 minutes the digital modulations switch off automatically.
Then I restart the ARB or I / Q generator (OPT 08) and the modulation can be heard again in the control revceiver.

Unfortunately, I do not have an RF spectrum analyzer to control.

Does anyone know this problem and can give good advice?

Thanks.
 

Offline Ordinaryman1971

  • Frequent Contributor
  • **
  • Posts: 320
  • Country: us
Did you check all the coax cables that are connecting modules? Almost impossible for those to disconnect because they are self locking but.. who knows.
All the LEDs that are reporting the power rails are lit too?
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Hello. I only received the device two days ago and I noticed this error during a first check. The case is sealed and I have not broken the seal yet.
Thank you.
« Last Edit: August 29, 2019, 05:32:28 pm by Jörg »
 

Offline Leo Bodnar

  • Frequent Contributor
  • **
  • Posts: 801
  • Country: gb
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Thanks, Leo. I have already read this article. Unfortunately, I could not see a clear match with my ARB error 518.
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Perhaps stating the obvious, but the error manual has this description:

518 Arbitrary Waveform Generator unlocked; refer to manual.  The
    instrument’s dual arbitrary waveform generator is unlocked. This
    error will occur when the ARB Reference key is set to Ext and
    there is no external signal applied to the instrument. It may also
    occur if there is a problem with the A5 dual arbitrary waveform
    generator board.

I would start by seeing if it will lock by giving it a valid external clock, applied to the BASEBAND GEN REF IN.  You will need to tell it the frequency you're providing.

Dual arb reference manual, if you don't already have it, which describes how to set up the external reference:

  http://literature.cdn.keysight.com/litweb/pdf/E4400-90328.pdf
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Hello MarkL. Thanks for the link.
In the large dockument library I only found the product note to the option UND.
I will read in the user guide how the external feed is done correctly.
 

Offline Ordinaryman1971

  • Frequent Contributor
  • **
  • Posts: 320
  • Country: us
Do you have board with OXCO?
Is the reference frequency set to external or internal?
Is there any offset set?
Any other errors... oven cold (only if you do have OXCO)?
How is the quality and integrity of the signal, is the frequency on spot? Is the output lvl ok?
Would you have access to power meter and/or spectrum analyzer?
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Hello.
No, the ESG 4433B has no OCXO option 1E5. Yes, reference set to intern.
The output frequency seems plausible from 100 kHz to 2 GHz. No noticeable amplitude fluctuations.

I took control with an Icom IC R 8500 as a control receiver.
Unfortunately I do not have access to RF Analyzer or Powermeter.

I am currently preparing the setup with GPSDO and frequency counter.
After warming up, I can say something about frequency stability and accuracy.
Then I do the test with the external reference for the AND generator.

No further error messages.
With the option UN8 (I / Q generator) sometimes "unleveled".
But this can also be a user error. I am still a beginner with the ESG and the complex settings.

I will contact you later with new results.
See you later.
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Now, I've checked and adjusted the 10 Mhz Ref. output of the ESG4433B.
The TCXO works well with a deviation of about 3mHz over 10Mhz.
Drift and ADev are not bad for a TCXO.
The amplitude of the generated signal seems to be correct. 50μV @ 15 MHz generate S9 on the S-meter of the ICOM.
Certainly, not a good replacement for a PowerMeter but a rough approximation.

I used the 10 MHz of my GPSDO as an external reference for the option UND (dual ARB).
I also checked that 10MHz is selected.

Changing between ARB Ref. Ext. or Int. shows no effect. For both sources, the error remains "518 ARB unlocked" and the modulation that I control with my receiver also remains completely unchanged.

Greetings
« Last Edit: August 30, 2019, 09:53:29 pm by Jörg »
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Sounds like a problem with the PLL on the UND board.

Ordinaryman1971 mentioned this already, but before digging further, can you confirm the REF IN connector on the board is connected properly to the BASEBAND GEN REF IN on the back that you just used for your test.  Sometimes people don't pay close attention to the connectors, if anyone before you worked on the insides of this unit.

What version of the board do you have?  It's either E4400-60187 or E4400-60069 and will be marked on the upper edge.

Understood you don't have a spectrum analyzer.  But maybe we can get by without one, especially if the problem really is the PLL.

Do you have a scope?  What model?  Does it have FFT?

I should also mention that before breaking any seals on the unit, if you only received it several days ago and you purchased it "Used" on ebay or from some other site as working, you might still have an option to return it as broken.  Once you break the seals, it's going to be a tough argument to make someone accept a return if you find something else wrong with it.  It's also not guaranteed the UND board can be fixed, if that feature is important to you.
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Hello MarkL.

Many thanks. Yes you understood correctly, I don`t have an RF spectrum analyzer.
I can only use the external audio interface to create the LF-FFT to about 80kHz.
 
The hardware information shows me that the DUAL ARB board has the version E4400-60187.

My scope is only analog with 60 MHz bandwidth. Not very helpful for that.

My guess is that switching the Ext and Int Reference for the ARB does not work correctly.
I can change the ARB Sampel Clock, so that changes the sound in the receiver.

What surprised me, is that I can switch the ARB reference to EXT without a message that no source is connected.

I will open the device in the next time.
Before that, I try to collect as much information as possible.
The digital modulations are not the most important thing for me at the ESG, but I'm just trying to make everything work correctly.

Greetings.
« Last Edit: August 31, 2019, 04:40:59 pm by Jörg »
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
I just tried to use the Agilent service software for the ESG.
I see  no possibility of control with the RS 232 port in the software.
Is there a trick for it or is the ESG only suitable for GPIB?
Jörg
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
I just tried to use the Agilent service software for the ESG.
I see  no possibility of control with the RS 232 port in the software.
Is there a trick for it or is the ESG only suitable for GPIB?
Jörg
It was years ago (at least 5), but I think I needed to use the HPIB port with a 82357B adapter.

I vaguely recall using the RS232 port to upgrade the ESG's firmware, but no other uses for RS232.
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
...
My scope is only analog with 60 MHz bandwidth. Not very helpful for that.
...

A 60Mhz analog scope will do.  The problem is likely in the PLL control circuitry, which is relatively slow-responding.  And the PLL clock output to the sequencer is no more than 40MHz anyway.

Since you don't have a spectrum analyzer, you won't be able to verify the RF output is actually correct.  But it should be possible to troubleshoot the PLL lock.

If you decide to proceed with opening the unit, here is a test you can perform on the PLL:

Set up the dual arb generator using MODE Mutli-tone..  The default is an 8-tone carrier with 10kHz spacing.  It will display temporarily it is creating a waveform.  Then turn off multi-tone.  You can turn on the RF or not.  It doesn't matter since the arb generator continues to run with the RF off.

Go back to MODE and enable dual-arb.  Choose Select Waveform and pick the waveform AUTOGEN_WAVEFORM.  This is the multi-tone waveform it just created.  Turn on dual-arb using the internal reference.  Loading the multi-tone waveform in dual arb mode allows you to play with the various clock settings, which you can't do in multi-tone mode.

Take a look at the following test points.  I have the same board as you, E4400-60187.  Note that testpoints (TP) are available on both sides of the board as pads.  You can probe them directly or solder small jumpers to them so you can clip your probes to them.

Here's what I see on my board with the above dual-arb waveform:

  TP200 PLL reference clock in = 312.5kHz, 0V to 5V square wave.

  TP201 PLL clock out = 312.5kHz, 0V to 5V square wave.

  J5 pin 2 PLL VCO tune = 8.48VDC, steady.

  TP204 PLL derived waveform clock = 5MHz, 0 to 4V square wave.  This should be the "ARB Sample Clock" setting divided by 2. (TP204 is near the middle of the board, far from the other TP2xx pads).

Your PLL clock out and tune voltage will probably not be right since the PLL is not locked.  The VCO tune voltage may also be oscillating as the PLL tries to lock, so look at it with a DVM and a scope.  The red LED next to U212 should also be lit on your board, indicating no PLL lock.

Try changing the ARB sample clock from 10MHz to something else, say 8MHz.  Do you see any change at TP204?

Try moving jumper J5 to the PLL test position, shorting pins 2 and 3 (normal is pins 1 to 2 (lower position)).  This applies a fixed 7.40VDC to the PLL VCO tune input (i.e., no feedback).  On my board, this results in approximately 469kHz at TP201.  The no lock LED should definitely be lit.
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
...
What surprised me, is that I can switch the ARB reference to EXT without a message that no source is connected.
...
The error might still be there.  You have to clear the error log to make it display again (under the "Utility" button).

If the error is persistent there should also be "UNLOCK" on the display.
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #16 on: September 01, 2019, 09:35:58 am »
Hello MarkL.
Your control measurement on the ARB board can be very useful.
Thank you very much, especially because no circuit diagrams are to be found.

I will measure the test points.
However, it will take some time. It's just a little bit of time for the hobby when I come from work.

Do you have experience with the replacement of the backup battery in the ESG.
Is a voltage from an external power supply required at the battery terminals, to prevent data loss?

Best regards.
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #17 on: September 01, 2019, 01:58:55 pm »
...
Do you have experience with the replacement of the backup battery in the ESG.
Is a voltage from an external power supply required at the battery terminals, to prevent data loss?
...
I have not replaced my battery.  Maybe I will now that the lid is off again.

The battery replacement procedure is described in the Service Guide under "A14BT1 Battery", page 5-24.  It states that only user data will be lost and not calibration data when the battery is removed.

I'd highly encourage sitting down and reading, or at least paging through, the available manuals and notes for this instrument.  It's all on keysight.com.  There's a lot of information.
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #18 on: September 07, 2019, 11:47:19 am »
Hello MarkL. Yesterday I opened the ESG.
I can say for the time being that all voltage control LEDs are lit on the mother board.
I have the UND board removed and cleaned. I've seen if I can localize smoked components. Unfortunately not noticeable. In a next session I will repeat your measurements on my ESG.

Do you still have your ESG open?
Can you tell which status LEDs are lit on your UND board? On my board lights a red and a green LED, no matter whether the dual ARB is turned on or not.

I'll get back. Thanks and Regards. Jörg
« Last Edit: September 07, 2019, 12:00:04 pm by Jörg »
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #19 on: September 07, 2019, 04:50:37 pm »
Unfortunately, I do not have my ESG open anymore.  I won't be able to get back to this for a week or so.

I'm fairly certain only one LED was lit in that row at the top, and it was green.  They're all driven by the large cell array chip next to them.  Maybe someone else can look, but I'm not sure what they'll tell you unless you found a description of them somewhere.

The red LED near the I/Q output SMB connectors is the PLL unlock, as I described previously.

When you get a chance to probe the PLL test points, feel free to also post pictures of the scope traces.  Sometimes subtle clues can be found by looking at the waveforms.

I found a better block diagram than the one I posted previously.  Extracted from the service guide.
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #20 on: October 12, 2019, 08:42:58 pm »
Hello Mark L, I hope you are still here.
I managed to do the control measurements on my ARB board today.

My comparison measurements with Juper J5 in normal position
TP200 PLL reference clock = 312.5 kHz, 0 V to 5.15 V square wave
(Clock present when ARB-ON, no clock when ARB-OFF)

TP201 PLL clock output = 415.7 kHz, 0 V to 5.15 V square wave.
(Clock always present when ARB-ON and ARB-OFF)

J5 pin 2-3 PLL VCO tune = 15.56 VDC, steady.

Waveform clock derived from TP204-PLL = 6.65 MHz, 0 to 8 V square wave.

The red LED next to U212 is always on. The Unlocked indicator on the LCD sometimes flickers.

I measured at Jumper J5:
Pin 1: 7.418 VDC
Pin 3: 15.56 VDC

My comparison measurements with jumper J5 in the PLL test position:
TP200 PLL reference clock = 312.5 kHz, 0 V to 5.15 V square wave

TP201 PLL clock output = 290.3 kHz, 0 V to 5.15 V square wave.
853726-0


Measurements on TP204 with Juper J5 in normal position:
Signal only present when ARB-ON
TP204 = 6.650 MHz (ARB sample clock 10 MHz)
TP204 = 3.325 MHz (ARB Sample Clock 5 MHz)
TP204 = 9.978 MHz (ARB Sample Clock 15 MHz)


Measurements on TP204 with Juper J5 in test position:
853730-1

As you suspected, VCO voltage and PLL output clock do not match.
The big problem for me is to find a mistake without proper circuit diagram in the SMD stuff.

Maybe you still have a good guess.

Many greetings jörg
« Last Edit: October 12, 2019, 09:13:44 pm by Jörg »
 

Offline JörgTopic starter

  • Regular Contributor
  • *
  • Posts: 80
  • Country: de
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #21 on: October 29, 2019, 05:43:48 am »
Hello. Now I have bought a used dual ARB board. The seller described it as tested and functional.
I had great hope that the problem with the Dual ARB UNLOCKED errors is solved.

Unfortunately, the error still exists with the new board.
I do not think that two ARB units from different series happen to show the same defect.
The error should have another cause.

Unfortunately, in the service manual I see  no description for the exact assignment of the connections to the mother board.
Also how the signals are correct at the upper connections,  I can not find.
In order to locate the defect, I need helpful information which signals are correct at the connectors.
Does anyone have more accurate information?
Thank you very much. Jörg
 

Offline thorswait

  • Contributor
  • Posts: 26
  • Country: ca
Re: Wanted, Help For Agilent ESG 44XXB Sig. Gen. - Dual ARB Error 518 Unlocked
« Reply #22 on: February 21, 2023, 07:57:30 pm »
Hi all,

I did spent a lot of time working on the E4400-60187 UND Dual Board with PLL issue. I did a lot a measurement but every time it brought me back to the U211A section. I did have 2 E4400-60187 B3904 Version 204 or 203 with the same issue. The PLL was locking when installed in my E4433B but when I installed on card in my E4436B it became intermittent ... and was not locking. I did replace about all the component in the loop without any improvement ...

I've found out that there is a flaw in the design the first comparator U211 is not isolate and can oscillate forcing the PLL not to lock. In order to fix this issue you need to lift the pin 3 of U211 (In +) and add a serie resistor and solder a wire to R212. I did put a 270K and it is working great on both board I've fix so far.

I hope it could help someone in the future.
 
The following users thanked this post: kazam, rplabs

Offline kazam

  • Regular Contributor
  • *
  • Posts: 73
I have this exact issue on my E4436B.

Did you mean to lift the pin and then add a 270k resistor down to the pad and then also add a wire?
 

Offline kazam

  • Regular Contributor
  • *
  • Posts: 73
Well, I did this and it didn't solve my issue unfortunately...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf