0

Number 057 mostly done... headphone amp issues
Moderators: altitude, adafruit_support_bill, adafruit, hamburgers, phono

Please be positive and constructive with your questions and comments.

Number 057 mostly done... headphone amp issues

by Exogenous on Tue Jun 21, 2005 1:52 am

First let me say how pleased I am that this project exists. 'b0x no. 057 is happily bleeping away.

Presently I get no sound from the headphone output in random mode. The line out plays fine. Pin 6 of the LA4140 (IC14), which by my reading of the schematic should approximate headphone out, reads +5.4 volts DC with negligible AC signal.

The input to this headamp chip (pin 2) shows a signal riding on about +2.5 volts DC offset. Is this normal?

I did notice a stray component lead in this area while getting things running, so something might have been damaged by a short. I also stupidly had one of the 74AC165 chips in backwards for some time but I'm not sure it would cause this problem.

I appreciate any suggestions for fixing this. I'm willing to get another LA4140 but am concerned about the DC at the input. When I get home late tomorrow night I'll have another go at sorting things out...
Exogenous
 
Posts: 24
Joined: Sat Mar 12, 2005 1:04 pm

by Guest on Tue Jun 21, 2005 3:17 am

does it only not work in random mode

you might have diode d33 in backwards
or a electrolytic cap in backwards
recheck all components and solder joints
Guest
 

by Exogenous on Sat Jun 25, 2005 9:24 pm

I've re-checked that the components are all in properly but there is still severe DC offset going into the headamp chip which seems to stop it from doing its job. I can't imagine how random mode vs. another mode would change this as the line out is fine.

I notice that there is some DC offset headed towards the line out but it is all blocked by C59. C51 is positioned to do the same for the headamp chip but substantial DC gets by.

My troubleshooting is hampered by my complete lack of understanding as to the functions of Q33 and Q34 and related passive components. Maybe I'll just install a cmoy for the headphone amplifier...
Exogenous
 
Posts: 24
Joined: Sat Mar 12, 2005 1:04 pm

by Guest on Sun Jun 26, 2005 6:47 am

Have you tried reloading the firmware? Give that a try.
Guest
 

by Exogenous on Sun Jun 26, 2005 11:22 am

It's inconceivable to me that the problem lies in the digital section. I think I will try swapping some of the transistors in case I fried the ones in the mixer section.
Exogenous
 
Posts: 24
Joined: Sat Mar 12, 2005 1:04 pm

by Guest on Sun Jun 26, 2005 3:15 pm

the transistors also run the mix out
so they are not the problem

have you checked your ribbon cable that supplies
the six volts and the headphone out
it could be in backwards

otherwise the la4140 is probably bad
Guest
 

by Guest on Sun Jun 26, 2005 4:50 pm

here is the voltages that should be seen on the la4140
set to random mode and volume full up
no headphones connected

pin1 1v offset 40mv of signal
pin2 3v offset 400mv of signal
pin3 3v offset 400mv of signal
pin4 3v offset 2v of signal
pin5 0v offset 0v of signal
pin6 3v offset 2v of signal
pin7 6v offset 0v of signal
pin8 6v offset 2v of signal
pin9 3v offset 0v of signal

these values are approximate
variations of 10% should be expected
Guest
 

by Exogenous on Sun Jun 26, 2005 8:41 pm

Thanks very much for measuring those voltages on the LA4140. It looks like the input of the LA4140 should be able to handle the DC offset. None of my pins where you found an amplified signal showed any signal for me. At this point I think you're right that my LA4140 is dead.

I found that Newark/MCM sells the LA4140. I've heard enough bad things about the outrageous shippings costs of Newark that I'm tempted to just wire in a cmoy, but I wanted a part (for another project) from them anyway... sigh...

Thanks though
Exogenous
 
Posts: 24
Joined: Sat Mar 12, 2005 1:04 pm

by Guest on Sun Jun 26, 2005 9:47 pm

sorry to hear about the la4140

before you build a new amplifier
check the voltages on both sides of c51 and r159
it sounds like the signal may not be coupling in well
if you find signal on the positive side of c51 try replacing it
same with r159
Guest
 

by Exogenous on Sun Jun 26, 2005 10:38 pm

C51 passes the signal. I temporarily replaced it with a smaller cap to better block the DC but still no signal coming out of the headamp chip.

I have a small extra cmoy (shown below) kicking around that should be easy to wire in, though it would certainly be neater to use the onboard LA4140.

Image
Exogenous
 
Posts: 24
Joined: Sat Mar 12, 2005 1:04 pm

by Milkmansound on Sun Jun 26, 2005 11:13 pm

my headphone amp does not work either - its not a big deal since I do not need to use it - but I do not want it to lead to other problems down the line. I guess I need to open the box up again and see whats happening in there - could it just be that some of the amplifier chips are faulty?
(((((MILKMANSOUND)))))
Milkmansound
 
Posts: 77
Joined: Wed May 25, 2005 12:33 am
Location: Brooklyn, NY

by Guest on Mon Jun 27, 2005 9:40 am

My guess is a faulty headphone amp section won't affect the rest of the machine, but it's worth seeing if any parts are getting hot and drawing too much current from the power supply. When I had a part in backwards early on, my voltage regulator heatsink got super hot.

If you want to debug the headphone amp, I suggesting looking for signal at some of the points as described in this thread (like C51). That's easier with an oscilloscope but will work with headphones as described in the fab manual. Use cheap headphones because you are likely to encounter some DC that could destroy them. I think I fried my LA4140 with a stray compononent lead that I missed when it landed across the chip sometime.
Guest
 

by Milkmansound on Mon Jun 27, 2005 2:08 pm

I dont even use the headphone portion, so maybe I will just desolder the voltage supply pin to that IC when I get a chance. Nothing heats up as far as I know - I did the finger test for at least 30 minutes during the initial burn in.
(((((MILKMANSOUND)))))
Milkmansound
 
Posts: 77
Joined: Wed May 25, 2005 12:33 am
Location: Brooklyn, NY

by adafruit on Mon Jun 27, 2005 5:47 pm

theres a list of 'odd transistor' sellers on the x0xb0x resource page: if you fried your la4140 (they -are- rather delicate) you can buy a replacement for pretty cheap from one of them (mcm/newark is bad, but maybe http://www.electronicsic.com/ isnt)

adafruit
 
Posts: 12151
Joined: Thu Apr 06, 2006 4:21 pm
Location: nyc

by Exogenous on Mon Jun 27, 2005 7:42 pm

Dang - I didn't think to check electronicsinc. Shipping & handling there shows as $7.99, which I regard as fair.

I just called Newark and it's too late to cancel... oh well, I needed a MC33340P from them anyway for another project. I'll report what the shipping damage is when I find out.
Exogenous
 
Posts: 24
Joined: Sat Mar 12, 2005 1:04 pm

Please be positive and constructive with your questions and comments.