Marconi/IFR/Aeroflex 2023 2024 2025 Repair advice, Error 502 Fractional N loop

[volvo_nut_v70]

I just bought a IFR 2025, it works just fine until it warms up. Then I get a ERR 502: Fractional-N Loop high.

It's has all the hallmarks of a thermal issue. Has anyone dealt with this before?

Many thanks in advance!!

[AlcidePiR2]

I got  a IFR2023a  with the same problem.

Err  502: Fractional-N loop  high

And no signal in the RF output.
I have checked the ref signal on the back BNC output
As soon as  the 2023a is turned on,  the 10Mhz  square signal appear, but very quickly  shut down.
See the video  (6 seconds)






You can hear the clicks of the relays.

One can see a  review of  the IFR2023a in the Eevblog #261   


But I must say, I have no clue where to search for this fault.

[AlcidePiR2]

I found some related and interesting information in the eevblog here

https://www.eevblog.com/forum/repair/quick-repair-project-marconi-2024/msg151502/#msg151502


and  to a lesser degree, here

https://www.eevblog.com/forum/testgear/looking-for-schematic-marconi-2031/msg452841/#msg452841

[AlcidePiR2]

Dear All,

I am starting a new attempt to fix  this beautiful device, but will need some help. I will really appreciate if someone could guide my steps, as the service manual exists, but is not always very explicit

http://bee.mif.pg.gda.pl/ciasteczkowypotwor/Marconi/Marconi_2023A_2024A_2025_Service_Manual.pdf

The problem is the  error message "Error 502 Fractional N loop"

This is what I have done so far, following the maintenance manual.

501,502 Fractional-N loop low, Fractional-N loop high.
These error messages can be caused by a number of faults on RF board AA1 or AA1/1.
The synthesizer section of the 2023, 2023A, 2023B, 2024 and 2025 is similar. However, the
variants are software-limited at different frequencies. For example, a 2023 is limited to
1.2 GHz output. When repairing a 2023 synthesizer entering exactly half the frequency
specified in the following section will have an identical effect. For instance instead of setting a
frequency of 1201 MHz, set a frequency of 600.5 MHz for a 2023.
The following tests can be performed without removing the covers:
(1) Step the carrier frequency from 1201 MHz to 2400 MHz (2510 MHz for a 2025) in 100 MHz
steps and note the frequencies at which the error is displayed. If the error is not present at
some of the frequencies it indicates correct operation of main dividers IC307 and IC309,
controller IC305, phase detector IC310, 311 and 312, and loop filter and lock detector IC402,
IC404 and IC406. If the error appears at all frequencies refer to the ‘General synthesizer
fault diagnosis’ section below.

Done :  the fault is present al all carrier frequencies . Thus

If the error appears at all frequencies refer to the ‘General synthesizer
fault diagnosis’ section below.

General synthesizer fault diagnosis
(1) Check that the voltage on TP14 is outside the range ±8 V. If not, check the operation of lock
window comparator IC406(c),(d).

Voltage at TP14 is 9.133V

[AlcidePiR2]

(2) Check that 25 MHz is present on IC205 pin 9 and that 5 MHz is present on pin 3 of IC313.
If not, refer to ‘Error 506 VCXO loop low’ or ‘507 VCXO loop high’ to debug the reference
loop.

- 25 Mhz is present on IC205 pin 9

- I think there is a misprint here.  pin 3 of IC313 has no 5 Mhz, but also no test pad.  pin 3 of IC310 has  a test pad and 5 Mhz is present.

Can someone confirm the misprint ?  Or am I wrong ?

[AlcidePiR2]

(2) Check that 25 MHz is present on IC205 pin 9 and that 5 MHz is present on pin 3 of IC313.
If not, refer to ‘Error 506 VCXO loop low’ or ‘507 VCXO loop high’ to debug the reference
loop.

- 25 Mhz is present on IC205 pin 9

- I think there is a misprint here.  pin 3 of IC313 has no 5 Mhz, but also no test pad.  pin 3 of IC310 has  a test pad and 5 Mhz is present.

Can someone confirm the misprint ?  Or am I wrong ?

I checked again and realized that the error was mine.

The numbering is confusing as the IC313 name is on the other side of the chassis separation. 



When measured at the pin 3 of IC313  we have a 5 Mhz signal with 5.2 V amplitude and  30 ns width.

[George Edmonds]

Hi

The first step is to check very carefully for leaking electrolyte SMD electrolytic capacitors, they are a PITA.

I have a full printed service manual if the PDF copy is not clear.

73 George G6HIG 

[AlcidePiR2]

The first step is to check very carefully for leaking electrolyte SMD electrolytic capacitors, they are a PITA.

Thanks for some feedback. This si supportive in this task. I have checked the SMD electrolytic capacitors. From what I can see visually,  they are fine.

I have a full printed service manual if the PDF copy is not clear.

This is good to know. What is the date of you manual ?

[George Edmonds]

Hi

May I first say that it does not help and adds to the confusion when what should be two separate threads become merged into one.

I intend addressing the problems with the one in France only in this posting, given, that it appears to be the simplest/quickest one to fault find.

The printed service manual that I have is dated 8th March 1999, exactly the same as the online version.

Errors in operators and service manuals are not unheard of, this covers all manufacturers, not just Marconi.  Your brain has to be kept active somehow.

Having re read the entire posting very carefully and if the information provided is accurate the initial problem with the one in France is clear.  It is stated that after a few seconds the 10MHz standard output signal on the rear of the instrument goes missing.  This is the first fault to be addressed, without a fully working 10MHz internal standard the instrument cannot possibly work correctly.

There are two types of 10MHz standard used in the 2023/2025 range of Marconi signal generators, a TCXO one made by HY-Q and a OCXO one made by NDK (option 4 on serial plate).  I have yet to see a faulty NDK standard but this is certainly not the case with the HY-Q standard which do fail.  BUT it must be made clear that the problem may not be one the 10MHz standards at all, they rely on the instruments power supply and the support circuitry on the AB1 control board all of which must also be checked and be fully functional.

Check that in the Util4 “frequency standard” menu you have option 4 selected, “internal 10MHz out”,

There are also two versions of the main instrument planar, which one is being used a AA1 or a AA1/1.

At risk of repeating myself, the main problem with these signal generators is the surface mount electrolytic capacitors failing, the smaller the physical diameter of them the more likely that they  are to have failed.  The smaller diameter ones appear to have problems with the electrolyte seals failing.

73 George G6HIG in Dover UK.

[AlcidePiR2]

May I first say that it does not help and adds to the confusion when what should be two separate threads become merged into one.

You may be right, but when I started this, I thought it was better not to multiply the threads on the same subject.  Next time, I will do a separate thread for a repair tentative.
Now it may be too late to change.

I intend addressing the problems with the one in France only in this posting, given, that it appears to be the simplest/quickest one to fault find.

Thanks a lot, your input is extremely valuable, as I am quite blind here.

Having re read the entire posting very carefully and if the information provided is accurate the initial problem with the one in France is clear.  It is stated that after a few seconds the 10MHz standard output signal on the rear of the instrument goes missing.  This is the first fault to be addressed, without a fully working 10MHz internal standard the instrument cannot possibly work correctly.

There are two types of 10MHz standard used in the 2023/2025 range of Marconi signal generators, a TCXO one made by HY-Q and a OCXO one made by NDK (option 4 on serial plate).  I have yet to see a faulty NDK standard but this is certainly not the case with the HY-Q standard which do fail.  BUT it must be made clear that the problem may not be one the 10MHz standards at all, they rely on the instruments power supply and the support circuitry on the AB1 control board all of which must also be checked and be fully functional.

Mine is HY-Q  ( TCO-1009 10.0Mhz 15/99)

Check that in the Util4 “frequency standard” menu you have option 4 selected, “internal 10MHz out”,

Thanks a lot George ! This is one of my mistakes. I had it set to 0. I set Util4 to 4, and now I have the  10 Mhz signal in the  rear output ( amplitude needs to be divided by 10
as I left the 10x option while using a simple BNC cable).

There are also two versions of the main instrument planar, which one is being used a AA1 or a AA1/1.

Where is this information ?  I cannot see it.

At risk of repeating myself, the main problem with these signal generators is the surface mount electrolytic capacitors failing, the smaller the physical diameter of them the more likely that they  are to have failed.  The smaller diameter ones appear to have problems with the electrolyte seals failing.

I understand. I will check carefully the  SMD caps this week-end.

Thanks again for the assistance. I hope that with your help I will be able to make this instrument run again.

[AlcidePiR2]

There are also two versions of the main instrument planar, which one is being used a AA1 or a AA1/1.

Where is this information ?  I cannot see it.

Now I have it

My board is  44830/001  so according to P. 6-4 of manual, it should be AA1  ( AA1/1 board is 44830/249).

[AlcidePiR2]

Let me continue my search for the fault, following the procedure of the maintenance manual .

(3) Using an oscilloscope examine the signal on TP7. A pulsed signal should be present with an
amplitude of 5 V and a pulse width of 25 to 50 ns.

In TP7, I find a pulsed signal, with frequency 10.66 Mhz, pulse width of 30 ns, and amplitude of 5.16 V ( see below) so this is OK.

(4) Using a frequency counter measure the frequency of the signal on TP7. If the signal has a
frequency close to 5 MHz (± 50 kHz), it is likely that the fault is within the phase detector or
loop filter components.

The frequency  is  10.66Mhz, so not in a condition of likely fault.

If the phase detector is operating, the voltage on pin 3 of IC402
should be at 5 V if the signal is above 5 MHz, or at 0 V if the signal is below 5 MHz.

The signal is at 10.66Mhz, thus above 5 Mhz, but the voltage on pin 3 of IC402 is  4.04. Is this Bad ?

If this
is correct, examine loop filter components IC402 and IC404 for correct operation.

How do I do that ?



Thanks

[AlcidePiR2]

At risk of repeating myself, the main problem with these signal generators is the surface mount electrolytic capacitors failing, the smaller the physical diameter of them the more likely that they  are to have failed.  The smaller diameter ones appear to have problems with the electrolyte seals failing.

I understand. I will check carefully the  SMD caps this week-end.

I have checked a dozen of the SMD electrolytics  ( desoldering it from the board and testing with a transistor tester ( 1.13k))

None is totally off

The 100 uF   and 47 uF have ESR of about 3R

The 10 uF have ESR of 12R  could this be a source of the problems ?

I will probably recap the whole thing to be sure.

[Miti]

I just measured some capacitors from my junk box:

Panasonic EB series 100uF 50V - 0.22R
Nichicom UHE series 100uF 50V - 0.025R
Nichicon UPW series 100uf 25V - 0.170R
Nichicon UPW series 47uf 25V - 0.638R
Nichicon UPW series 10uf 25V - 0.648R

So I think your caps are a bit high. How many hours does it have? I have an IFR2025  with 77492 hours so almost 9 years of continuous operation. 

[AlcidePiR2]

The caps  are slightly lower voltage

10 uF   16 V
47 uF   16 V
100uF    6.3 V

But I agree that the ESR is too high. I just wonder if this is sufficient to cause the problem. In any case, I will change all caps.

I had not check the hours of service of the device. Now its too late, as I will not power it without the caps on.  Will look  after putting back the caps.

[AlcidePiR2]

This unit has 78493 h of operation

I have changed  all electrolytics from the RF board, and now it is working.

I am surprised as none was really off, as I have seen in previous repairs. But obviously, they were the culprits.
I wonder also if there was not a bad joint at one of these caps.

Thanks for all !