Topics

Escaping smoke #ubitx

Bill KC5SB
 

Got my ubitx wired up today and it was working great!  Receiving and transmitting well.  I decided to try a different speaker and in doing so, somehow I must have shorted the speaker leads, resulting in all the magic smoke escaping from U1 and R75 burned in two.  Aargh!

Got some new parts on order - hope this did not take out anything else.

Jerry Gaffke
 

Very curious that R75 would smoke,
that's pulling low an input pin to the unused 2'ed channel of the TDA2822M.


On Sat, Feb 24, 2018 at 03:11 pm, Bill KC5SB wrote:
Got my ubitx wired up today and it was working great!  Receiving and transmitting well.  I decided to try a different speaker and in doing so, somehow I must have shorted the speaker leads, resulting in all the magic smoke escaping from U1 and R75 burned in two.  Aargh!

John
 

Unfortunate I had the same experience also smoked U1. Will give a good look at R75. It blew my 2A fuse.
First day I just had a Bluetooth speaker connected via a 3.5mm cable. I had no power on SSB so was ready to start researching yesterday. Plugged in a Sony headset and white smoke appeared.
Ordered parts too, unfortunate also a long delivery time.

John

John
 

I used my phone to look closely at U1 and was shocked about the soldering. So guess while waiting for parts I can check all soldering and create a case.

Bill KC5SB
 

It happened pretty fast.  I tried to get the power shut off quickly but was not fast enough.  The smoke was coming from U1 and it was very hot to the touch.  I thought I might have caught it in time - made sure the speaker leads were not shorted and powered it up again (I know, not smart).  This time I saw more smoke and a spark before I got it shut down. I should have examined everything closely before powering up again.  I think the spark might have been R75 in its death throes.  All other surrounding components look OK.

Doug W
 

On Sat, Feb 24, 2018 at 03:37 pm, John wrote:
I used my phone to look closely at U1 and was shocked about the soldering.
Did you look at the other side?

Jerry Gaffke
 

I suspect that solder job is just fine, 
Have you gotten around to looking at those pins from the bottom of the board?

The surface mount parts are autostuffed.
The through-hole parts are then hand-added.
All you need is a good fillet of solder connecting pin to pad on the back of the board
where the hand soldering job was carried out,


On Sat, Feb 24, 2018 at 03:37 pm, John wrote:
I used my phone to look closely at U1 and was shocked about the soldering. So guess while waiting for parts I can check all soldering and create a case.

Dexter N Muir
 

On Sat, Feb 24, 2018 at 03:43 pm, Bill KC5SB wrote:
- made sure the speaker leads were not shorted
... and not earthed? others have reported frying chips ... :)

Bill KC5SB
 

Yep, one of the speaker leads could have been grounded - not sure.  Whatever happened, U1 came out on the short end of the stick.  Note to self - be careful with speaker leads.
Examination of the solder joints on the bottom of the board indicate no problems and quality workmanship.

Jerry Gaffke
 

Not clear why this is happening, but:

The TDA2822M is rated for operation with a supply between 1.8v and 15v, which is an amazing range,
and suggests the output drivers are pretty low impedance.
Since we are running it at 12v (or more), which is near the top of its range, a shorted output could indeed be all it takes.
Specs and graphs in the datasheet stop at a 9v supply, it's only the spec for the supply voltage that 
suggests 15v might be ok.  

If you run the TDA2822M output to a stereo headphone jack with both ears getting sound,
then plugging in a mono jack will short the output to ground. 

If you wind up replacing the TDA2822M, you might lift pin 1 and insert a 4 or 8 ohm 1/2 watt resistor in series there as a precaution.

If really paranoid, perhaps add an 8 or 9v linear regulator (or string of diodes) to reduce Vcc into the TDA2822M.

I have no idea how R75 could have smoked, except maybe a catastrophic failure in the TDA2822M 
shorted that (normally) input pin to Vcc somehow.

How many have blown?   Maybe 2 or 3?  In all cases, is it possible the output got shorted?

Jerry, KE7ER


On Sat, Feb 24, 2018 at 04:54 pm, Bill KC5SB wrote:
Yep, one of the speaker leads could have been grounded - not sure.  Whatever happened, U1 came out on the short end of the stick.  Note to self - be careful with speaker leads.
Examination of the solder joints on the bottom of the board indicate no problems and quality workmanship.

Joe Puma
 

Glad I saw this. You wouldn’t think you would smoke something by shorting the speaker leads. That can easily be done by plugging in a headphone jack or not having the headphone jack wired correctly the first time. Be careful! ;) 


Joe



On Feb 24, 2018, at 7:54 PM, Bill KC5SB <blmerideth@...> wrote:

Yep, one of the speaker leads could have been grounded - not sure.  Whatever happened, U1 came out on the short end of the stick.  Note to self - be careful with speaker leads.
Examination of the solder joints on the bottom of the board indicate no problems and quality workmanship.

John
 

Mine might have shorted when I plugged the headphones in. 

I did also check the bottom of the board and that looked fine. Guess there is a bus than placed in the board to connect the under and upper copper plane? Otherwise I would think that some solder on the upper side would connect that trace a little better.

John

Jerry Gaffke
 

The uBitx and Bitx40 are built on a quality FR4 fiberglass-epoxy pc board with plated through holes.
Check the mounting holes in the four corners, you can see a metal cylinder fully connecting the top and bottom pads.
That's true of all holes on this board.
No need to solder pins to more than one side of the board.


On Sat, Feb 24, 2018 at 05:28 pm, John wrote:
Mine might have shorted when I plugged the headphones in. 

I did also check the bottom of the board and that looked fine. Guess there is a bus than placed in the board to connect the under and upper copper plane? Otherwise I would think that some solder on the upper side would connect that trace a little better.

William R Maxwell
 

My thoughts too, Jerry. I suspect that the uBITX is many builders' first exposure to double sided boards with plated through holes.

Bill, VK7MX



On 25/02/2018 11:03 AM, Jerry Gaffke via Groups.Io wrote:
I suspect that solder job is just fine, 
Have you gotten around to looking at those pins from the bottom of the board?

The surface mount parts are autostuffed.
The through-hole parts are then hand-added.
All you need is a good fillet of solder connecting pin to pad on the back of the board
where the hand soldering job was carried out,

On Sat, Feb 24, 2018 at 03:37 pm, John wrote:
I used my phone to look closely at U1 and was shocked about the soldering. So guess while waiting for parts I can check all soldering and create a case.

Bill KC5SB
 

I ordered 5 TDA2822M IC's with sockets and several 2.2 ohm resistors for R75.  I intend to use the IC sockets. 
Tomorrow I'm going to pull U1 and R75 from the board and use a signal tracer to see if I am getting audio into this stage.  I think I should be able to pick off the signal at the wiper of the volume control (?).  Is there any risk to doing this?

After the parts are replaced I plan to use the bench supply to bring the voltage up slowly and carefully watch and monitor temperature after I replace U1 and R75.  I was fiddling with speaker jack and don't remember the exact sequence of events before the catastrophic failure.  I will report back after the parts come in.  Thanks for the replies guys.

Arv Evans
 

Many of us "old timers" are used to one lead of the speaker being at ground.  This does not work with newer audio amplifier ic,s that have differential output.  Neither side of the output can be grounded...not even if DC isolated by a capacitor.  

Arv
_-_




Sent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------
From: "Jerry Gaffke via Groups.Io" <jgaffke@...>
Date: 2/24/18 6:09 PM (GMT-07:00)
To: BITX20@groups.io
Subject: Re: [BITX20] Escaping smoke #ubitx

Not clear why this is happening, but:

The TDA2822M is rated for operation with a supply between 1.8v and 15v, which is an amazing range,
and suggests the output drivers are pretty low impedance.
Since we are running it at 12v (or more), which is near the top of its range, a shorted output could indeed be all it takes.
Specs and graphs in the datasheet stop at a 9v supply, it's only the spec for the supply voltage that 
suggests 15v might be ok.  

If you run the TDA2822M output to a stereo headphone jack with both ears getting sound,
then plugging in a mono jack will short the output to ground. 

If you wind up replacing the TDA2822M, you might lift pin 1 and insert a 4 or 8 ohm 1/2 watt resistor in series there as a precaution.

If really paranoid, perhaps add an 8 or 9v linear regulator (or string of diodes) to reduce Vcc into the TDA2822M.

I have no idea how R75 could have smoked, except maybe a catastrophic failure in the TDA2822M 
shorted that (normally) input pin to Vcc somehow.

How many have blown?   Maybe 2 or 3?  In all cases, is it possible the output got shorted?

Jerry, KE7ER


On Sat, Feb 24, 2018 at 04:54 pm, Bill KC5SB wrote:
Yep, one of the speaker leads could have been grounded - not sure.  Whatever happened, U1 came out on the short end of the stick.  Note to self - be careful with speaker leads.
Examination of the solder joints on the bottom of the board indicate no problems and quality workmanship.

Jerry Gaffke
 

You don't have to be all that old.

Both the BItx40 and the uBitx have one lead of the speaker at ground.
It's only when you ground the wrong lead on the uBitx that things get "interesting".

A needed clarification, as I can hear 1000 forum readers 
scrambling to get the soldering iron out.

Jerry


On Sun, Feb 25, 2018 at 08:30 am, Arv Evans wrote:
Many of us "old timers" are used to one lead of the speaker being at ground.  This does not work with newer audio amplifier ic,s that have differential output.  Neither side of the output can be grounded...not even if DC isolated by a capacitor.  
 

Bill KC5SB
 

I did some post-mortem examination on the audio output stage today after U1 and R75 went up in smoke.  Here is what I found:

As Jerry pointed out above, plugging in a mono jack to the ‘phones’ jack will short the output to ground.  I believe that this is exactly what I did.  I tried to plug in an extension speaker with a mono jack I’m pretty sure that that is when the failure happened.

 I was able to carefully remove U1 and there appears to be no damage to the board or any traces.  I also removed what was left of R75 and cleaned up the board with alcohol.  Interestingly, a pad on one side of R75 appeared to have little or no solder. This was the side of R75 attached to pin 6 of U1 (input 2 +).

 

 I carefully applied power to the unit – all seems well except no audio output of course.  Transmit on SSB and CW are working.  I could hear a modulated SSB signal on another receiver which sounded OK.  I probed the connection to U1 pin 7 (input 1 +) with an oscilloscope and was able to see a signal.  I then tuned the rig to 14.074 (20 meter FT8) and sure enough, every 15 seconds the signal would briefly disappear and then reappear for another 15 seconds.  This indicates to me that the previous audio stage is working – here’s hoping anyway.  I thought about injecting this into an external audio amp but decided not to push my luck – I’m just going to wait for the parts to arrive and work on a case in the meantime.

 

Did not mean to ramble so much, but decided to post this in case it might help someone else. Incidentally, I probably qualify as an old timer – first licensed in 1960.  Still learning something every day and making rookie mistakes!

 

Bill

 

 

Bill KC5SB
 

Installed the replacement parts today - back up and running.  Going to work on the case and a few mods now.  Thanks for the help everyone.

Bill

Spldrong@...
 

okay, so i am very new to all of this. I just finished wiring up my uBITX and I plugged in a Mono speaker to test it. everything was working fine, except i noticed that the volume was not loud at all. 
    Then the audio just shut off. Everything else is still working.
    So is this probably because I used a mono speaker with a mono jack? and now I need to replace my TDA2822M?

Thank you.