The damage wasn't hard to spot. Q17 has overheated so badly the case is
charred and cracked. C54 is burned beyond recognition, with a matching
smoke spot on the inside of the top cover.
The schematic shows Q17 as a 2SA1359 PNP transistor. The emitter is
connected directly to the 13.5 volt power input, right after the
reverse-polarity protection diode. C54 is between the collector of Q17
and ground. I'm leaning toward the theory that C54 shorted and caused
excessive current in Q17, but I don't know these radios well enough to
know if that makes sense, or if I'm seeing the results of a more serious
failure elsewhere (like a bad PA module.) It's a bit hard to suss out
what Q17 does precisely because no pinout is provided for the PA module.
Anyone have more experience with this rig?
The damage wasn't hard to spot. Q17 has overheated so badly the case is
charred and cracked. C54 is burned beyond recognition, with a matching
smoke spot on the inside of the top cover.
<cut>
_________________
Service manual is here if you need it:-
http://www.mods.dk/manual.php?brand=icom
Colin
** The service manual is here:
http://www.radioamateur.eu/schemi/IC28_serv_WA0KKE.pdf
Q17 switches the output level from the RF power amplifier module SC1019 ( a
custom Icom part ) from 5 watts to 25 watts. Almost any low voltage, general
purpose PNP transistor in the same pack & capable of a few amps current will
do to replace Q17 - eg BD436 / 438.
http://www.fairchildsemi.com/ds/BD/BD434.pdf
http://www.toshiba.com/taec/components2/Datasheet_Sync//66/7578.pdf
If the failed electro really was the cause of Q17 failing, then the unit
should work after replacing it and Q17.
However, if Q17 still overheats then suspect the SC1019 module is blown.
... Phil
Thanks, Colin. I did find that, and it's been very helpful. I also
realized the block diagram had transistor numbers in it, which tells me
that Q17 is part of the automatic power control circuit, which would
explain the lack of power output.
I may just go ahead and change C54 and Q17 and see what happens.
Thanks, Phil. That's also the conclusion I've come to, so it's nice
to get a bit of a sanity check that I'm not overlooking something
obvious. :)
Right now I'm working under the theory that the SC1019 module is OK,
because I otherwise can't explain why C54 would be burned to a crisp.
A bad PA module could cause excessive current through Q17 but it
shouldn't have affected that capacitor.
I'll let you know what happens when I put it back together.
Thanks for the circuit rundown. I followed along on the block diagram
and it's a lot clearer to me now.
I don't actually think it was reverse polarity, because there's a
large polarity-protection diode that should have caused the input fuse
to blow before any damage could be done. (Hmm, maybe I should check
that diode and make sure it's not open, though...)
My current working theory is that C54 developed an internal short
(maybe due to age) and cause excessive current in Q17. Seems like
it's pretty much got to be that or a bad PA module, and a bad PA
wouldn't explain why C54 burned. I've never seen a capacitor so
thoroughly immolated. It left a big smoke spot on the inside of the
case, and it just fell apart when I poked at it with a screwdriver.
My current working theory is that C54 developed an internal short
(maybe due to age) and cause excessive current in Q17. Seems like
it's pretty much got to be that or a bad PA module, and a bad PA
wouldn't explain why C54 burned. I've never seen a capacitor so
thoroughly immolated. It left a big smoke spot on the inside of the
case, and it just fell apart when I poked at it with a screwdriver.
** Is C54 actually a * tantalum* bead cap ????
If so, no special explanation is needed for its fiery demise.
... Phil
Yeah C54 destroying like that is a bit of an enigma. Check the crowbar
diode. Some people like to do stupid things when they blow a fuse like
wrap it in foil or replace it with a 20 amp. The radio could have been
powered up without a proper fuse and unattended when the failure
occurred. That would have caused a cascading catastrophic failure.
I've seen this happen so many times I've lost faith in my fellow hams. :(
If you have some way to limit current, use it. I have a supply that I can
dial in from 0 to 35 amps and it comes in real handy
Could be. It's 10 uF, polarized, which could be either. It was too
badly charred to tell what it used to be, but I didn't find the
remains of an electrolytic can rattling around in the case, so a
tantalum does seem likely.
Will do. I don't have a variable-limit supply, but I do have one with
a fixed 1.5A limit that I plan to use for initial low power transmit
testing.
"Phil Allison"
> ** Is C54 actually a * tantalum* bead cap ????
>
> If so, no special explanation is needed for its fiery demise.
Could be. It's 10 uF, polarized, which could be either. It was too
badly charred to tell what it used to be, but I didn't find the
remains of an electrolytic can rattling around in the case, so a
tantalum does seem likely.
** FYI - tantalum bead caps look like this:
http://www.allproducts.com/manufacture97/suntanbe/product1.jpg
They are known to catch fire and emit smoke and bits of the tantalum oxide
"slug" for no good reason.
If you found a "can" then the cap is an ordinary electro that overheated,
the electrolyte boiled and burst the seal.
... Phil
That will help you not fry some of the bigger stuff but there is nothing
like a 0 to 15 volt, 0 to 35 amp supply with volt/ammeter.
Even better if it has OCP. If the current drain reaches a user programmed
setting (IE those BJTs are about to blow!), the power supply will shutdown
its output completely, instead of just dumping continuous current to
whatever the current is set to.
I have an Agilent 6643A which does this. While they cost $3K new
you can pick 'em up for a 10% of that on Ebay. I paid just a bit over $300
for mine.
That's great if you know within a decent tolerance how much current what
you are working on draws at inrush or at an idle.
True. Start low and work your way up. :)
And put your nose close to it and inhale. I do a lot of troubleshoot with
my senses. After 30 some years I can distinguish just what's hot or
burning, smell ozone from HV, hear things snap crackle pop etc...
Also a good tip, but when doing so I wear safety glasses. I've had a few
things go critical mass just as I lowered my nose for a sniff! Gotta love
those - practically fell out of my seat.
How true :) I've had my fair share of electrolytic out-gassing up the
nose. Never had anything with the potential to injure blow up in my face.
Didn't find a can, so almost certainly it's a tantalum. Good catch, i
probably would have replaced it with an electrolytic if you hadn't
said anything.
Hoping to get this thing put back together later this week. Been way
too busy to get to it 'till now. I'll let you know what happens.
Didn't find a can,
** Funny how you used that word earlier.
so almost certainly it's a tantalum. Good catch, i
probably would have replaced it with an electrolytic if you hadn't
said anything.
** Did you read the warning above about tants I just gave you ???
There is no reason to use them if space permits a normal electro.
.... Phil
OK, fair enough. I wasn't sure if they had different leakage
characteristics or something like that.
I replaced Q17 and C54.
I now have correct voltages on all four pins of the PA module in
transmit mode, but still almost no power output -- I suspect a bad PA
module, at this point. Reinforcing this is the fact that I appear to
have more signal on the input of the PA module than I have on the
output.
I think I'm going to set this one aside and just use it as a receiver;
at this point it's not worth spending much more time on, unless I run
across another IC-28A with a different problem.
Your assumption is most likely correct... I can give you an idea how
ever.. If you can remove the cover of the module, and inspect the board,
you may find a crack. You'll need a high temp soldering iron to put it
back together.. This is common with RF modules or various types.
I've done this 3 times already on different rigs and it worked. Most
of the time a crack develops, mostly from abuse. Other times, you have
a bad module..
>Your assumption is most likely correct... I can give you an idea how
>ever.. If you can remove the cover of the module, and inspect the board,
>you may find a crack. You'll need a high temp soldering iron to put it
>back together.. This is common with RF modules or various types.
>
>I've done this 3 times already on different rigs and it worked. Most
>of the time a crack develops, mostly from abuse. Other times, you have
>a bad module..
It doesn't require abuse. See:
<http://www.11junk.com/jeffl/pics/drivel/slides/power-amp.html>
That's the guts of the VHF power module from my Kenwood TM-732A radio.
The circled areas are where I had to solder the trace back together.
The two failures were seperated by about 2 years and will probably
continue at the same interval. My guess(tm) is that it was caused by
either incomplete baking of the hybrid, or a bad solder paste mix when
the hybrid was silk screened.
The gaps were almost invisible and very difficult to see. They would
have been far more visible if I could have shined a light from the
back, but there's a big copper heat spreader in the way. Instead, I
found them with an ohmmeter. I started with the probes at the end of
the conductor trace, which showed no continuity. I slowly moved the
probes closer to each other and eventually closed in on the break.
Soldering was also rather difficult. The ceramic hybrid is a
wonderful heatsink and literally sucks all the heat out of the
soldering iron. I had to use a rather large 850C screwdriver solder
tip and still managed to do a lousy job of soldering.
Also note on the photograph the two big power xsistors. The goo
covering them is rather soft and flexible silicon rubber. Please
resist the temptation to push on it, or you'll break the wire bonds
underneath.
If desperate, you can get replacement PA modules from RF Parts.
--
Jeff Liebermann je...@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
No new PA available or is it too costly?
I use a mini gas torch that has nice tips for soldering, no flames get
in there but the tip gets hot as hell. The heat sink has no chance of
holding back. :)