Icom IC-736 front panel problem

[mehdi]

Today I came across an Icom IC-736 which seemed very very new (the seller claimed it wasn't used and I think he was right)
I turned it on. It turns on, the LCD has backlight, but is empty (no text is shown on the display) and a white noise can be heard from the speaker; S-meter also is on its maximum value (right-most). The panel buttons don't work.
I checked these things with no success:

• Reset the CPU (holding Ent and CLEAR and turning on the device)
• Took the lithium battery out
• Checked the power supply voltage at some test points and they were OK
• Checked the voltage of some logic board ICs and they were 5V

R20 (47 ohm) on control panel unit seemed to have problems (had changed color a bit) but it was OK (the 7805 besides it was also OK)

Any ideas what could be wrong, or how can I pinpoint the problem?

Thanks in advance

[TheSteve]

If all test voltages from the service manual are fine then I would carefully re-seat all of the various board interconnects. If that doesn't make a difference then we'll dig into the service manual further.

[Radio Tech]

I would check those connectors as TheSteve requested.
Also remember this radio has a built in power supply. And it is a switch mode unit.  Check the rails for correct voltage and ripple.

[mehdi]

Thanks.
All the voltages are OK; The CPU does not send any control commands to the LCD driver;
I think one of these could be the reason:
1- The CPU may have been damaged.
2- The EPROM contents may be corrupted.

What do you think? (We have another IC-736, so we can copy the contents of EPROM and program the faulty one with it)

[GandalfDerGraue]

any signals on the CPU to be seen with an oscilloscope? If not, is there kind of reset logic nearby the CPU, some capacitor being pulled up or down by some resistor going directly to an CPU Pin? Often there are problems with dried electrolytic caps.

[AF6LJ]

The radio can be remote controlled by computer  I would see if I could get the radio to handshake with a computer, this would rule out the CPU...
I am assuming some front panel functionality...

Also can you put the radio into transmit with ether the microphone or the front panel transmit switch??

Have yuou put an antenna on the radio and checked for signals??

[Nuno_pt]

Check the CPU's on the back of the display and check all the voltages, I've seem lot's or radios with no display cause by this.

Also I've see an 1000mp with flickering display and no audio, and the problem turn out to be one LM7509, that was not supplying the correct voltage to the dispaly CPU. 

Other times can be the light of the display, like on the 1000mp xxx series and all with CCFL light the light could be bad, and there is no light, so no numbers, but the display is working.

[njvprodux]

Hi, did you sort out your Icom 736 because I killed mine with exact same symptoms and have fixed it. On my radio I was playing around with an external antenna tuner plugged into the tuner socket - stupidly I started packing up and disconnecting with the radio still on and the display just went blank and it fell silent - exactly as you describe. I checked everything simple - no luck. I traced where the external tuner wires go - unbelievably straight onto the big 5v CMOS I/O chip without any protection or buffering. I figured I had shorted the 13.8v line to one of the tuner 5v control lines.

I actually bought another 736 to act as reference and help with repairing mine so was able to compare chip voltages, 'scope signals etc. The CPU was running but in a half second repetition waveform - I figured it was starting, hanging, resetting in an endless loop.

I built one of those simple component analysers that put out low voltage AC and display the current drawn on a 'scope - this showed the pins that the tuner connects to on the I/O chip were very different to the reference radio - I checked every pin and found several anomalies. Assuming my analysis of the destruction was correct I couldn't see any other chip that would be damaged by the 13.8v so put my money on the I/O chip being bad. The problem was always going to be that the protection diodes on these chips can steer the high current 13.8v onto the 5v rail or any data line so almost any other chip on that board could possibly be destroyed as well.

So I managed to buy a couple of replacement I/O chips TMP82C265BF-2 and today did the repair. Very disappointed when it failed to get the radio working. For some reason before packing up for the day I decided to try swapping the only chip that is socketed - the EPROM from the reference 736 - Bingo - it worked! So, on my radio the I/O chip was blown by 13.8v from the tuner socket and that must have gotten onto the CPU's address or data bus and took out the EPROM as well. Everything now works, at least I haven't yet found anything that doesn't. The EPROM seems to be a bog standard 27C256 device so will have a go copying the reference and burning a new one tomorrow.

I hope this is helpful
Best regards
Paul
GK3NJV

[AF6LJ]

Let us know how that turns out.
Thanks for passing that along.

[njvprodux]

Hi, yes no problem at all - I had some 27C256's in my junk box from so very long ago - ! They were programmed and the problem these days is no one has a UV eraser any more. A friend had a programmer so I tried putting the chips out in the sun for a week or so. When I went over to my friend he confirmed (1) the EPROM from the faulty radio was bad (2) the EPROM from the reference radio was good and he took a copy of the contents (3) one of the EPROMs exposed to sunlight was blank, the other wasn't and (4) the code from the reference radio went into the blank EPROM and this then worked inside the repaired IC-736 radio. RESULT!

It's taken me 18 months to repair this radio, I was in no hurry having bought another one. What I was afraid of was that I had a faulty radio that might be repairable - but if I screwed up removing the bad I/O chip - it's an 80 pin quad package - and damaged the PCB, I'd just have a dead radio! Period.

I bought some Chipqwik solder and flux plus a hot air blower from ebay. I set everything up very methodical and carefully. I screened everything off around the faulty chip with aluminium foil and made a little fence around the chip using thin aluminium cut from a sardine tin. I wound some fine wire through the legs of the bad I/O chip. With my air blower mounted in a drill stand on minimum temperature setting (100C) I placed 4 pieces of Chipqwik, one on each side of the chip. I added a good amount of the Chipqwik liquid flux, switched on and waited. I saw the solder melt - you have to be patient, it takes 3 to 5 minutes for the Chipqwik solder to eat into the lead free stuff originally used - try tiny touches with tweezers - mine took just over 3 minutes and then - amazing - the chip moved! I was able to lift it off the PCB. Using cotton buds it was easy to wipe all the excess solder into blobs and get them with my solder sucker. A few minutes work had it all removed. The PCB cleaned up with an Isopropyl Alcohol wipe also in the Chipqwik kit - the result was absolutely perfect - the PC looked like new. I was amazed how good this worked and how easy it was after all my worrying. This was 9/10ths of the job done - getting the old chip off without damaging the tiny fragile PCB pads.

Fitting the new chip was a case of very accurately aligning it with the PCB pads and then tack soldering the corners. Once this is done gently run the iron tip with fresh solder down each row of pins until all pins are flooded with solder and shorted together.  Then use fresh "solder-wick" to soak up the excess solder - it comes off easily and leaves a perfect amount behind to solder the chip. Another clean up with Isopropyl alcohol and a brush/tissue and very careful examination to detect any remaining pin short circuits and the job is done. I was my first attempt at SMD work and the result is perfect - you simply cannot tell any work was done at all.

There are plenty of videos on Youtube that show these processes.
73 Paul

[AF6LJ]

Good Deal...