Aeroflex IFR-2398 / calibration / repair help

[tech255]

zed dez.

thanks for the advice not sure if your aware but the 2399 has an extra rf lead to the top of the synth module and as there is no manual available I have no idea what it does, my friend recently dropped off his anritsu spectrum analyser for me to use to find the faults as it goes out to 6ghz.
I will have a look at the yig at the weekend and see if I can see anything loose, not hopeful on finding spares.
have tried to find a company in the uk that can handle repairs but no joy, its such a nice machine would hate to scrap it.

cheers

[George Edmonds]

Hi all

The 2398 is a LG (Korea) re badged product, not an IFR/Marconi.  I believe there is NO support for them now, what there was was a module changing service and the faulty modules went back to LG for repair.

What was Marconi/IFR/Aeroflex/Cobham, is now Viavi solutions Ltd who are only interested in 5G cellular equipment

I have the service manual for a 2399 which should be similar, but it is a module change manual and probably not much help.

73 George G6HIG

[Addie]

Hi All.  Did anyone locate a copy of FCAL2398 - cal backup/restore utility?  Thanks!

[vk2cow]

Hi all,

I have acquired a faulty IFR2398 and after some searching online I found this old thread. The symptoms are similar to those described by others (level fail, the built in cal signal is -42 dBm and external signals are over 20 dB less than expected). I suspect at least one fault at the front end as the SWR looking into the input is 4:1 or worse, which is really unlikely for an instrument like this. Has there been any other development other than those described in this thread?

cheers
Dimitris

[George Edmonds]

 LG 2398 006.pdf (165.2 kB - downloaded 232 times.)

Does this help

G Edmonds

[vk2cow]

Thanks George, this looks like the circuit in my analyser indeed. I am guessing this is from the service manual of the LG version of the same instrument?

[vk2cow]

For the benefit of future generations troubleshooting an IFR 2398.

Initially:

0) All self cal pass exept the Level cal

1) I got SWR 4:1 or worse (antenna analyser looking into spec an input, spec an switched on, SWR the same regardless of attenuation selection (0 to 50 db, 10 dB step).

2) I see the spec an built in cal 40 MHz signal at -42 dBm (should be -30 dBm) and any signal provided to the input of the spec an was over 20 dB less than what it should be.

My thinking:

3) I focused to the circuit directly at the input of the spec an (as indicated by the SWR behaviour). Thanks to George for the schematic.

My actions (so far):

4) I supplied -12 volts to the board, then measured the voltage on each of the two control pins of every AH002R2 RF switches. All of them had ~-8.8V and 0V but the first one had -1V and 0V.

5) I removed the first AH002R2 switch

6) I connected directly with a short jumper to connect the N-type input to the rest of the attenuator chain (pin1 to pin7 from memory).

7) Re-installed the RF module. All self CAL failes exept the YIG - expected as there's no connection to the 40 MHz internal cal signal. SWR is now 1:1 (yey!!!)

Just for fun, added an external 40 MHz about -30 dBm signal to the input of the analyser and rebooted it. It passed all tests! Under these conditions, it behaves as you would expect but the amplitude seems to be out by ~3 dB. Given the dodgyness of the calibration, that's hardly a surprise.

I have ordered a AH002R2 from eBay (fingers crossed) and will report what happens when I install it.

cheers
Dimitris

[doarmin]

Bonjour rien depuis fevrier dernier y a t il du nouveau depuis le remplacement du circuit??
Personnellement je recherche le schemas de la carte vidéo detecteur IFR2398 ref 513-648A.car comportement bizarre autour des capas chimiques
C175 à C177.

[George Edmonds]

Hi

These electrolytic capacitors and a none electrolytic capacitor, plus a number of resistors and an inductor appear to be form just an input filter for the +15VDC supply to the board.

G Edmonds

[George Edmonds]

Hi

Please note that I cannot and will not provide full schematics, here is a partial scan showing the +15VDC input filter.

G Edmonds

[doarmin]

Bonsoir merci beaucoup celà m'aide énormément...J'avais bien vu que ces condensateurs étaient en parallèle mais cette platine a au moins trois couches
et pas simple de suivre...J'ai un peu moins de 6 ohms  aux bornes de ces condensateurs et de nombreuses résistances de 15 ohms  R01 à R08 se sont coupées face à ce presque court circuit une seule semble avoir tenu le coup ...celà explique la présence encore du +15V venant du connecteur de l'alimentation principale.
Résultat : pas de vidéo et échec au niveau du level lors de l'autotest.par contre la Fi 10,7mhz rentre bien sur cette platine  visiblement l'option peack n'est pas concernée par ce + 15V en presque court circuit mais d'autres circuits sur la platine doivent etre concernés mais j'ai pas de détails....ou alors un des condensateurs C173 à C177 défectueux...J'aimerais bien car travail de fourmis autrement.
Bests regarts.

[George Edmonds]

Looking at the information that I have it appears that the +15VDC supply to virtually every IC on the board is decoupled with a 22uf 20V working capacitor with either a 0.1uf or a 0.01uf capacitor in parallel, some IC’s look like they us the +15VDC twice with separate decoupling capacitors.

Regrettably there is no power distribution schematic or a parts list in the information that I have.  May I suggest that you measure the resistance of the +15VDC supply to ground either side of the two 22 Ohms resistors in parallel, this may give you a clue as to if the fault is the filter or is one of the IC supply decoupling capacitors.

All the best and hope that you succeed, if I can help more just ask.

George Edmonds

[doarmin]

Bonsoir.
Faisant suite à vos conseils j'ailevé une à une les condensateurs du filtre puis les découplages du +15V je retrouve encore 2?5ohms par rapport à la masse.
Plus de doute il reste trois CI reliés à cette ligne +15V
U5  U16 et U13 .Plus rien de connecté sur cette platine.A part U5 bien connu un CA2111A les autres ne me disent rien...j'ai besoin d'aide avant de les retirer l'un après l'autre...

[George Edmonds]

Hi

Before I continue may I say that I only have a very limited understanding of technical French and am basically using Google Translator, so certain points are probably being lost in translation.  I also do not have an IFR2398 at present so I am working from an incomplete set of factory schematics.  I do have the IFR service manual, but it is a module replace only manual as all boards/modules were intended to be returned to the factory in South Korea for low level repairs to be carried out

In your last email you said “U5, U16, and U13 Nothing else connected on this plate”.  I suspect that plate should be PCB, from what information that I have there are some 38 IC’s on the PCB with well over 50% using the +15VDC supply, this supply is also used for multiple other purposes and is heavily decoupled with electrolytic capacitor using 0.1uf capacitors in parallel and sometimes just the 0.1uf capacitors.

I do not believe that removing capacitors/IC’s one after another is a valid method of finding the fault as you are very likely to cause damage to the PCB by doing so, my suspicion is that the PCB is at least a four layer one.

In the past I have asked you to check to ground the resistance at either end of the R223 in parallel with R223A.  These resistance measurements should be made using a four wire Kelvin setup.

How you find the short on the PCB depends on the equipment available to you, you need to be able to put a constant current into the PCB +15VDC supply rail and then trace the test current to the fault, there are many ways of doing this, but they all involve specialist equipment that you may not have.  I have an HP set of probes that can inject a pulsed current and then follow the pulsed current about the PCB to the shorted device.

I have scanned the 511-648-A board block diagram and have attached it to this posting.

George Edmonds

[doarmin]

bonsoir merci pour votre réponse.Je vois que vous etes très équipé en matière de recherche de panne...Il y a certes beaucoup de circuits intégrés sur la carte mais après avoir cherché les connections au +15v il y en a seulement 3 qui le sont directement les autres sont à un 5V distribué autrement ou alors il y a une régulation permettant peut etre de fabriquer un +9 10 12 ou autre mais n'ayant pas le schemas...
S'est effectivement une carte avec plusieurs couches et le +15V est distribué ainsi sur toute la carte...En connection directe il y a les trois ci que j'ai identifié après un vrai jeu de pistes:
 Un CA2111A. un MC1350.  et un CA3089.
Oui s'est trés risqué sur un multicouches de sortir un ci traversant comme le CA2111A.En fait la nuit dernière j'ai eu une idée : comme la résistance entre le OV et le +15V est de 2,5ohms j'ai alimenté la carte avec 2V puis utilisé une camera Thermique ce qui m'a permis de voir C154 chauffer puisque parcoutu par un courant de 800mA...méthode certes un peu hors normes mais le résultat est là.Ce condensateur pourtant vérifié etait passé hors vigilance ayant du vernis sur les soudures...L'IFR2398 est reparti comme avant.merci pour votre aide dans cette aventure car j'aurais eu du mal à trouver sans vos conseils.

[vanderleySCL]

Hello everybody,
My IFR 2398 has a problem on the RF input board, specifically the capacitor C1 is leaking, I couldn't get an electrical schematic to check its capacitance, does anyone know how to tell me the value of this capacitor?

[vanderleySCL]

I installed a 10N capacitor (103) while not finding the correct value and it seems to work fine. But I still have another problem which is an error in LOG calibration. This problem seems to be in the attenuator because the REF LEVEL when above 10Dbm as measurements are correct but if I change to 9.9DBM (or any value below 10DBM) as measurements are incorrect, the signal a drastic reduction. If anyone has the attenuator schematic and can share it with me.

[George Edmonds]

Hi

C1 is a 1mfd none polar capacitor 25vdc working.

I posted the schematic of the attenuator previously, look back at my earlier postings.

Looks like this 2398 has had a DC input overload.

G Edmonds

[vanderleySCL]

Thank you very much Edmonds, I will provide the correct value of the capacitor, you are right, it suffered an overload on the input, the error problem in the Log I was able to identify, it is a 20DB attenuator relay that is damaged (Teledyne a150-20-12) , I made an external attenuator just to test and it passes the self test, now I need to buy this relay. Very grateful for the support.

[EB4CUV]

Hi George

I bought an IFR 2398 with a fault and I'm seeing that the input step is blown. In a previous message you said that you had posted the schematic of this input board but I can't find it. Could you send it again or tell me where to find it?

Thanks, Julio.

[George Edmonds]

H

Reoly #54 in this posting.

G Edmondsi

[EB4CUV]

Hi George,

Sorry for the inconvenience. In that post #54 the attached .pdf is from the rceiver front-end board and I don't see any other archive.
The schematic that I need is of the input board where the N connector is and not for the front-end because the diodes are burned.

Thanks,
Julio

[George Edmonds]

Hi Julio

Having now had a little time to look at your photograph I would be very surprised if this is the only damage.

At present I cannot provide a schematic of the input assembly, but you do not need one, just a little information.  The only RF path components are C1, DC blocking capacitor which is a 1uf at 25V device and the two catcher diodes CR1 and CR2, these diodes are MSMP-3892, data sheet attached.

All the rest of the input assembly is providing a plus and minus reference supply derived from the main instruments + and – 15VDC supply.  CR3 and CR4 are MMBZ5232LT1 Zener diodes.

Regrettably I suspect that there will be further downstream damage, particularly to the solid state switches.

Hope this helps

G Edmonds G6HIG

[EB4CUV]

Hi George,

Thanks for your help George. People like you make life easier for others.

I've removed the bad diodes from the input board, leaving just the capacitor until a replacement arrives.

Your suspicions are true. The first solid state switch on the front end board is damaged.
I have done what vk2cow said in post #56, remove the first AH002R2 switch and bridge pins 1 and 7.

Now, I inject a 40Mhz and -30db input signal into the N connector and the device passes all the tests correctly.
I'm on the right track.

By injecting other signals, the device measures almost correctly. The next step is to do a good calibration.

The next question is: where can I find the calibration software or some other that can handle the device?

Again, thank you very much for your time and help. I hope my experience is useful to others as well.

Julio, EB4CUV

[Ron Castle]

Hi George,

Ours is working but the center of the screen is all washed out. I guess I can use the video out would you know what the display P# might be.