Batboard

On the save issue, I'll see if it's a known bug. Could you get me the version # of the unit's software?

As for the handle info - I'll have to say "contact Service".

A newer DS1260-100 arrived from China today (2008 code date). I reinstalled it and all is well.

The 1600 did not like the 2 diode and battery solution at all.

Hello Wowbagger,

The firmware version is 4.02.

I do have another question. I just ran through the Spectrum Analyzer calibration through the ANT port and note that I now have it tracking really well from 0dBm to -90dBm.

However , if for example I set a level of -40dBm and note that it is dead on on the ANT port and then move the signal and switch to the T/R port the displayed signal level is +4dBm high.

My test setup is as follows:

Wavetek 3001 set to CW at 500MHz at -30dBm. The Wavetek output is connected through a low loss cable to a precision 10dB attenuator. The output of the signal at the 10dB attenuator has been verified with a Boonton 4200 Microwattmeter and further confirmed with an HP8566B both recently calibrated so I know that the signal level is correct.

I did not find any adjustments that were related to the displayed signal level in the T/R mode in the service manual.

All of the other calibrations appear to be okay.

Any ideas?

Thanks,

Sam

I did a little more investigating and found the the T/R input is 3dB higher than the ANT input as displayed. I verified that the power term path from the T/R input to the output that goes to the IF is 40dB down. Injecting a signal past this point also shows 3dB high. I did notice that the path from the ANT in jack to the receive IF has a 3dB pad in series with it where the power term path does not.

So I was wondering if there is an adjustment somewhere to compensate for this or is this normal?

I did see two pots on the Receive IF assembly but no indication in the service manual as to what they adjust. It appears to me if I put a SMA 3dB pad on the output of the power term where it feeds the IF I could compensate for the difference. However I would have thought that IFR would have noticed this and compensated for it somewhere.

Thank you,

Sam

Hello Wowbagger,

Well I was working on the1600S tonight calibrating the power meter. When I exited the routine and answered Y to the back up cal data when I restarted the system all I get is a beep every second and a blank screen.

Any suggestions?

Sam

I searched the Aeroflex site for any information re the 2945...

a) software update/upgrade to latest version
b) purchase of a software option
c) purchase of service/cal manual
d) purchase of parts

...but didn't find the slightest scrap of info. Is all that "secret" or am I missing something? (I'm used to Agilent equipment & support, website, etc.)

Also, what are the differences in the 2945, 2945A, and 2945B? Were they re-designs due to parts availability issues, or...? And, what' are the differences in the 2947/A and 2948/A/B variants?

Thanks, and a HUGE thanks to Wowbagger for contributing here!

The 294X family was designed by what was the old Marconi Instruments group in the UK, so I'll have to ask over there.

Software upgrade and purchasing of options and manuals should be handled though Service, just like anything else - I know they can handle that line.

As for the differences... I can ask the UK guys when I have one of my teleconferences with them, but I cannot guarantee an answer.

I have a ifr 3902 and I have several dstar repeater systems and radio. At one time I read that icom and aeroflex was going to come out with a option for the icom dstar systems, anyone know when and how much?

aj4g wrote:I have a ifr 3902 and I have several dstar repeater systems and radio. At one time I read that icom and aeroflex was going to come out with a option for the icom dstar systems, anyone know when and how much?

The D*star stuff is basically a set of canned waveforms IIRC - it doesn't do dynamic simulation (i.e. you won't be talking into the mike and hearing it).

I'd suggest contacting Service for pricing info.

Co6aka wrote: Also, what are the differences in the 2945, 2945A, and 2945B? Were they re-designs due to parts availability issues, or...? And, what' are the differences in the 2947/A and 2948/A/B variants?

I came up with nothing - I'd suggest asking Service.

Wowbagger wrote:I came up with nothing - I'd suggest asking Service.

Service didn't know either. In fact, the tech I spoke with seemed to know very little about 2945-series. didn't have manuals handy, and couldn't help me with ordering a system software update nor an option add-on. Claimed he'd be checking in to things and would call me back, but as of yet no call...

Hey,

Just started using an old COM-120A a few days ago to tune some duplexers. Worked perfectly until we went to get some coffee and when we came back it was stuck at the IFR boot logo screen stating something to the effect of "REINITIALIZING CALS TO FRONT PANEL"...not sure about the exact wording (I'll double-check it when I have a second). Anyone know what's up with that?

The calibration data is very important, so it stored in 2 places: the main CPU's nonvolatile RAM, and the EEPROM on the front panel's microcontroller. Each copy is checksummed.

At start up, the system evaluates the checksum on both copies. If one copy has a bad checksum and the other a good checksum, the bad copy gets overwritten with the good copy and a message is displayed. If both copies have valid checksums and match, then all is well. (if both copies have valid checksums but don't match, use the version in the main CPU until/unless told to load the front panel from the calibration screen, in case the CPU or front panel have been swapped).

So either your front panel got corrupted, or the front panel EEPROM is bad, and the only copy of the calibration table is in NVRAM.

Is there something we can do to try & reset it, or is it a support/repair case?

Thanks for your quick reply.

Don't worry about it - my personal unit's been doing it for some time.

Wowbagger wrote:Don't worry about it - my personal unit's been doing it for some time.

That's the thing - the message is stuck on the display, is there a key sequence we should be doing to bypass this?

Any key will do.

Hello.

Dear David (Wowbagger), probably you will be able to help me consultation, council, I will be grateful to you for any help.
(I apologize for my bad English)

Is faulty the spectrum analyzer IFR AN920 (options 02,08,09 are included).
The analyzer turns on, one beep, a bit later two beeps sounds, there is a picture and further, seconds in 20-25 seconds triple melodious beep and everything as the initial picture with the IFR logo was indeed, the instrument doesn't boot further.
Advise please, what triple melodious beep, an error means?
Whether it is possible to learn, define at what stage there is an error?
And already proceeding from it to look for failure in the specific unit further.

Very much I hope for your council as not to find the service description or the diagram.

Yours faithfully, Alexander.
Callsign – UA1ACV
Russia, St. Petersburg

The system is not finding the actual main application. Check the flash memory card on the main CPU board - see if it has come loose. If reseating the card does resolve the problem, then the card has become corrupted. Contact Service and get them to send you another card.

Hello David (Wowbagger)!

Thanks for the response.
Specify please. As far as I understand, it is the memory stick MELCARD Mitsubishi 1Mb F-EEPROM CARD MF81M1-GBDAT01 (a sticker on F000-0210-102 AN920 VER.1.02 card), it is set on a processor board under the unit of the display, the PCMCIA connector, on it is stored a firmware or the software which glitched, truly?
Reinstallation of the memory stick didn't yield result. If generally to disconnect it, the spectrum analyzer turns on, but there is no image on the monitor and pictures, beeps too aren't present.
I need to order such card in Support desk, correctly?

Thanks.

Yours faithfully, Alexander.
Callsign – UA1ACV
Russia, St. Petersburg

You are correct about what component is the memory card. It contains the application. The fact the system boots at all indicates the firmware is OK. The card is corrupt - the section that contains the program for the display is valid, but the section that contains the main analyzer program has been glitched.

Again: Contact Aeroflex Support. Get a new flash card from them.

Hello David (Wowbagger)!

Thank you for a valuable advice and consultation.
I wrote the letter to the official distributor in Russia AEROFLEX (IFR, Marconi), they are engaged also in service, technical support of products of AEROFLEX.
With request about memory card provision, conditions, delivery time.
http://aeroflex-cdip.ru

Yours faithfully, Alexander.
Callsign – UA1ACV
Russia, St. Petersburg

Wowbagger wrote:Any key will do.

About our COM-120, after further testing, unit is unresponsive to any keypress. Anything we could check ourselves? Or do we need to send it in?

Johnny Grep wrote:

Wowbagger wrote:Any key will do.

About our COM-120, after further testing, unit is unresponsive to any keypress. Anything we could check ourselves? Or do we need to send it in?

So, the screen is up, but it the stops responding to keys? It may be a bad connection on to CPU; my unit does this sometimes. Unseat the CPU from the socket and research it, or try just pressing in into the socket.

Hello,

I have an IFR COM120A that has a screen that is simply completely illuminated orange. It is quite bright with no text at all as the entire screen is lite up. Is this a common fault that someone could point me the direction of the problem? I would love to get a copy of the Maintenance Manual but they seem very hard to find.

I would be grateful for any thoughts from the forum.

Regards,

Trevor

hello yachtbird,I had a COM-120B in 2011,that year I tried to repair the EL screen due to some display problem,not very success.But I had found datasheet of that EL screen,and it said 2 potentiometers under back of screen,one is adjust brightness and theother I forgot function(also about display),I think you can try to find datasheet by model of you display,and try to turn it by manual.
Now I have 120A,it also have problem-----about celibration,selftest,options......I need more help.

Wowbagger wrote:You are correct about what component is the memory card. It contains the application. The fact the system boots at all indicates the firmware is OK. The card is corrupt - the section that contains the program for the display is valid, but the section that contains the main analyzer program has been glitched.

Again: Contact Aeroflex Support. Get a new flash card from them.

hello Wowbagger,could you tell me if someone people can change the celibration password of COM-120A/B/C?

Hello bh7juo,

I haven't returned to the problem because I am currently overseas but I do know what I want to do if I get the time. I really need to know if the display is faulty or the signals driving the panel is faulty.

I have had a look at signals on the connector to the panel. One of the signals is about what I would expect as the horizontal repetition rate. Another of the signals is what I would expect as the vertical scanning rate.

Other signals are hard to decipher but they may be because the intensity may be modulated by a 4 bit word or something like that. Some various voltages are also on the connector.

I would like to:

Trigger horizontal scans on a CRO with the horizontal rate. Trigger a sawtooth function generator and feed that to the input of the CRO. Now we have a raster like a TV set. Now I want to take one of the bits used for video and make it modulate the intensity of the CRO beam. If the video data is the Most Significant Bit it will provide some form of picture on the CRO screen.

This is a lot of trouble but I need to know if the panel has the problem or the driver circuits and I figure this can eliminate one or the other.

No circuit makes it really difficult.

Regards,

Trevor

hello Wowbagger,could you tell me if someone people can change the celibration password of COM-120A/B/C?[/quote]

No. It's hard coded into the firmware. Why would you want to change the cal password anyway?

yachtbird wrote:Hello bh7juo,

I haven't returned to the problem because I am currently overseas but I do know what I want to do if I get the time. I really need to know if the display is faulty or the signals driving the panel is faulty.

I have had a look at signals on the connector to the panel. One of the signals is about what I would expect as the horizontal repetition rate. Another of the signals is what I would expect as the vertical scanning rate.

Other signals are hard to decipher but they may be because the intensity may be modulated by a 4 bit word or something like that. Some various voltages are also on the connector.

I would like to:

Trigger horizontal scans on a CRO with the horizontal rate. Trigger a sawtooth function generator and feed that to the input of the CRO. Now we have a raster like a TV set. Now I want to take one of the bits used for video and make it modulate the intensity of the CRO beam. If the video data is the Most Significant Bit it will provide some form of picture on the CRO screen.

This is a lot of trouble but I need to know if the panel has the problem or the driver circuits and I figure this can eliminate one or the other.

No circuit makes it really difficult.

Regards,

Trevor

That should work, within limits. The first engineering prototypes of the video system did exactly that to drive an EGA monitor - the video timing parameters in the video controller had to be different to make EGA timings, vs. the EL timings. However, you are going to have a problem in that the EL panel timings have almost no horizontal retrace delay, nor any vertical retrace, as there's no electron gun + big magnetic field to sweep back.

Hello Rob61,

No, the IFR COM120A does not have an output like the IFR1600s. I have a COM120A with a display problem and that is one of the frustrations as I cannot see any status to guide me.

I do have a question for the group however.

I have now decided the fault is not with the display but rather back in the Digital Tray or somewhere else. On the Digital tray two RF signals are connected by rigid coax. One is the 1Mhz ref signal (W28). It is a beautiful 1 Mhz signal. The other is a 700Khz signal (W6) connected to the Receiver Tray. This signal is missing. My question is does this signal originate from the Digital Tray or does it originate from the Receiver Tray? Has anyone come across this before?

Many thanks in anticipation and as always any suggestions very gratefully received.

Regards,

Trevor

In regard to my previous post I now realise the 700 Khz signal provides a signal from the Receiver Tray to the Digital tray to drive the RF error meter. That is not my problem.

I am still on the case trying to work out why the EL screen is bright orange. It seems no video content from Digital Tray so I am looking for a 64 pin plug and socket so I can make an extender and operate the Tray out in the open.

Trevor

yachtbird wrote:In regard to my previous post I now realise the 700 Khz signal provides a signal from the Receiver Tray to the Digital tray to drive the RF error meter. That is not my problem.

I am still on the case trying to work out why the EL screen is bright orange. It seems no video content from Digital Tray so I am looking for a 64 pin plug and socket so I can make an extender and operate the Tray out in the open.

Trevor

I have checked IFR COM-120B's EL diaplay 14-PIN definition.Most important 4 signal are:PIN13----VIDEO DATA,PIN11----CLOCK(20MHZ),PIN7----VS(about 20KHZ???forgot),PIN9----HS(60HZ).I tried input VIDEO DATA,CLOCK and another signal which a little different form HS(from VIDEO BOARD a pad beside chips),to VGA monitor(auto mode),I seen display OK,but displayed resolution can't recognised by monitor.So if you check signal from these pins all seens OK(use oscilloscope),maybe EL panel bad.

The digital card is in the first slot so you can remove the front of the 120 to get full access.

You can remove the EL from the front and lay on top but given you have no useful information on the screen would just leave it disconnected for the moment.

I would make sure you have
1. removed all the aluminum caps
2. clean the board with alcohol and tooth brush.
3. Inspect the board for any damaged traces and verify continuity
4. replace capacitors
5. test with unit assembled and verify failure
6. remove front to gain access to the digital card and use the info from the last post to troubleshoot

The backplane connectors used are standard but not common ( i forget the designation used for that series ). The double row ones can be extended with ribbon cable type connectors but the triple row requires a PCB or other technique

jry and Bh7juo,

Many thanks for that advice.

Firstly brilliant to suggest taking off the front panel. I have been looking at this thing for so long sometimes you need a prompt for a simple solution like that. I have already ordered the connectors but don't need them now. The front panel is removed and I have great access to the digital tray.

Also, many thanks for the important signals on the 14 pin connector. I previously resolved that the EL panel is ok because when I inject a square wave onto Pin 13 I get a random response on the EL panel so I figured it was responding to video input. On pin 13 I have a constant 5V DC level. It has a little noise on it but it is not a video signal.

I was also concerned that I could not see any VS (approx 70 Hz) anywhere on that connector. I can see the HS signal ( approx 27 Khz) on pin 9. I thought it was supposed to be on pin 7 but your note confirmed that so thanks. I have no vertical sync signal. On pin 7 it is 0v.

The graphics chip (TMS 34010) is generating Vertical and Horizontal sync. The 40Mhz oscillator and 20Mhz oscillator are working. The video clock signal (20Mhz) and the input clock signal (40Mhz) are present on the graphics chip.

I will trace the Vertical sync signal as it is the obvious missing signal on the 14 pin connector to the EL panel. I think it may pass through a tri-state driver so I am investigating that next.

Tough without a circuit but I like a challenge. My fear is that one of the programmable chips is faulty but I figure that is a pretty remote case so I am not contemplating that at this stage.

I will replace the electrolytics as a matter of course just to cover that off.

Many thanks for your advice. VERY much appreciated.

Regards,

Trevor

Hello..

I have a COM120B that just came up with "Real Time Clock Bad" The units tx/rx seems to be working but I have no traces on the Analyzer or Scope....

Is there a battery, beside the 3.7 one on the controller tray? any ideas where I should look further???

Thanks!

There is a Dallas RTC chip on the processor board next to the DRAM battery ( more or less )

Is has an internal battery but not sure that would cause the issue you are seeing but it's possible.

I have replaced these before and they have to be programmed as well ...usually install a socket and a dab of Silicon RTV just in case.

Hello jry,

Can I please confirm something about your advice? You say remove all aluminium electrolytics. Later you say replace all capacitors. Do you mean replace all electrolytic capacitors or do you mean replace ALL capacitors including all of the small chip SMD capacitors?

I am not sure if the capacitor problem on these units is related to electrolytic capacitors or if it refers to all of the capacitors including the little chip SMD capacitors used everywhere.

Many thanks,

Trevor

Thanks!! Found it "Dallas Watchdog" and from what I have read the battery is good for about 5 years......Soo. how do you program it once I replace it?

on the digital card repair just the aluminum electrolytics should need to be replaced. The issue is that they will eventually fail if they are not leaking and defective now.

On the Watch Dog timer chip replacement on the processor it takes a $600 XLTEK programmer ( at least that what I spent ) that supports the chip. Chances are your old one has lost the memory and that data has to be reconstructed as well.
Normally the factory service would have to do this but it appears they no longer support these products. You can call IFR to see if they will handle for you or PM me if that does not work out.

yachtbird wrote:jry and Bh7juo,

Many thanks for that advice.

Firstly brilliant to suggest taking off the front panel. I have been looking at this thing for so long sometimes you need a prompt for a simple solution like that. I have already ordered the connectors but don't need them now. The front panel is removed and I have great access to the digital tray.

Also, many thanks for the important signals on the 14 pin connector. I previously resolved that the EL panel is ok because when I inject a square wave onto Pin 13 I get a random response on the EL panel so I figured it was responding to video input. On pin 13 I have a constant 5V DC level. It has a little noise on it but it is not a video signal.

I was also concerned that I could not see any VS (approx 70 Hz) anywhere on that connector. I can see the HS signal ( approx 27 Khz) on pin 9. I thought it was supposed to be on pin 7 but your note confirmed that so thanks. I have no vertical sync signal. On pin 7 it is 0v.

The graphics chip (TMS 34010) is generating Vertical and Horizontal sync. The 40Mhz oscillator and 20Mhz oscillator are working. The video clock signal (20Mhz) and the input clock signal (40Mhz) are present on the graphics chip.

I will trace the Vertical sync signal as it is the obvious missing signal on the 14 pin connector to the EL panel. I think it may pass through a tri-state driver so I am investigating that next.

Tough without a circuit but I like a challenge. My fear is that one of the programmable chips is faulty but I figure that is a pretty remote case so I am not contemplating that at this stage.

I will replace the electrolytics as a matter of course just to cover that off.

Many thanks for your advice. VERY much appreciated.

Regards,

Trevor

I remenber that before 4 important signals feed to connecter,there is a buffer chip(like 74HCT374 or etc.)to buffer signals.My 120B had failured it,so video data and sync signal lost.Of course I replaced it and worked OK after.I suggest you check this buffer chip first.

After much signal tracing I note the blanking signal is not present at the output of the TMS 34010 graphics chip. Without the blanking signal, I think the serial clock on the VRAM is inhibited (As it is missing from one of the FPGAs that generate the serial clock signal) and so it seems to me the internal I/O registers on the TMS chip are not being loaded and the TMS chip does not know the timing parameters of the graphic system and cannot provide a valid Blanking signal. Horizontal sync and vertical sync are present but I guess they may default to a value in the absence of instruction from the internal registers.

I figure the TMS34010 gets loaded with this information in the first couple of milliseconds from switch on but this may not be happening. I guess the EPROMS are either faulty or some other reason exists for the registers not being loaded.

All assumptions of course. Any ideas always appreciated.

Trevor

Thanks jry, I did a "replace the battery mod" but as you said it has probably lost it's program.... I will give IFR a call and PM you if they cannot help out.... Again Thank You!

Hello Wowbagger,

After much research I am taken back to one of your original comments, "My first thing to check would be to reseat those EPROMs - it sounds like the graphics processor is not coming up.". I am now in the realm of software analysis and not quite sure how to progress.

Is there any way I can see if the graphics processor registers have been loaded? I cannot see any blanking signal from the Graphics chip (horiz and vert sync ok) and I am guessing that is established after the registers are initialised. Any tell tale indication would be helpful if that exists?

As always, any advice appreciated.

Trevor

That's going to be tough - the GPU doesn't really have any signal it's working other than the display coming up.

Thanks so much for your offer. I am contemplating a purchase of either a 1900 or 1600 service monitor. The data sheet for the 1900 indicates the low cutoff frequency is 10mhz. However, two users I have talked to routinely use the 1900 down to 455khz. Could you tell me if there is 1900 build that officially is calibrated down below 10mhz (455khz). I plan to use the equipment for repair/servicing and alignment of HA, VHF, and UHF radio equipment.

Thanks for your assistance,
Bob

Wowbagger wrote:That's going to be tough - the GPU doesn't really have any signal it's working other than the display coming up.

Many thanks for the reply. I have found what I thought was a faulty LCC Eprom socket so I replaced that but to no avail. I have no blanking signal from the TMS 34010 graphics chip. I have horizontal and vertical sync but no blanking. The blanking signal I think enables the read clocks for the VRAM so I am trying to nut that out.

I have an eprom programmer but it does not have the socket to take the small LCC version of the AM27C512 eproms used in this unit. Before I obtain the test socket, does anyone know the checksum of both AM27C512 eproms on the digital tray that load the TMS34010 GFX chip? If I cannot find the checksums I will not try to get the appropriate socket for the programmer as that will be a waste of time.

Any advice always appreciated,

Regards,

Trevor

Thank you, Wowbagger, for your reply to my message. I will post my question here for all to see. I have just acquired a 1600S from a lab on the west coast. It is functioning with a certificate of calibration. I am using it for HF work mostly. I noticed that when using the receiver, I do not see a mode setting for LSB and USB only BFO. Also, the soft key selections do not match what is shown in the operators manual. This unit has firmware version 4.02 and options 20, 21, 35 and 36. I ran into a similar problem with a 1900. With the NAMPS option installed, for whatever reason, the receiver SSB functions were completely disabled .
Could you shed any light on this condition and if the LSB/USB functions can be restored on the 1600S
Thanks for your time.

Hello, I have a COM-120B, the problem is that, when testing a radio's frequency, it is giving a lecture and suddenly it gives another different and lower lecture, could the problem be in the reciever tray?

KN4HH wrote:I noticed that when using the receiver, I do not see a mode setting for LSB and USB only BFO. Also, the soft key selections do not match what is shown in the operators manual. This unit has firmware version 4.02 and options 20, 21, 35 and 36. I ran into a similar problem with a 1900. With the NAMPS option installed, for whatever reason, the receiver SSB functions were completely disabled .

It's a question of IF filters. There's only one position for an optional IF; both SSB and NAMPs require optional IF filters and so cannot be installed in the same box.