Maker Pro
Maker Pro

LM380N cuts out with loud input....sometimes

hrtendrup

Dec 20, 2013
2
Joined
Dec 20, 2013
Messages
2
disclaimers:
1. TLDR warning (don't bother unless you're bored)
2. I've built my circuit as a new cook would follow a recipe. I don't always know why I'm doing what I'm doing, but the recipe (schematic) says so, so......


About a year ago I build a headphone amp for my guitar as directed in this book. To my surprise, it worked rather nicely. To amuse myself, I installed the jacks, pot, and perfboard inside a washed plastic juice can, made a label for it, and called it guitar juice. I was quite pleased with myself.

This year, I had the idea to build five more of these things and give them to some of my guitar playing friends and family. I still had most of the passive components, but I needed to buy more LM380N ICs. I got 10 more from Jameco. I built the circuits, installed the ICs in the sockets and........intermittence.

Some of the circuits seemed to work and some did not. Then, when rocking/replacing the IC the ones that didn't work started working (until they stopped). Ones that had worked stopped working after a 2nd verification attempt. After some trial and error, mostly error, I concluded that after I got the circut working and I could hear output in the headphones, the circuit would be fine until I hit the strings hard on the guitar. I don't have a tone generator, so a plugged in guitar had to do. Likewise, if I cranked down the level/volume on the guitar itself (maybe half way, maybe 3/4, I don't know), it didn't matter how hard I hit the strings. So, it was, apparently related to the level of the input signal.

I put away the soldering iron, and sat down with just the multimeter and one of the circuits. Now would be a good time to check the attachment (that I'm likely not supposed to post). I checked that there was no resistance to everywhere I thought there shouldn't be and validated that there was some level of resistance where I thought there should be. I.E. I did the best I could to validate I didn't have any solder bridges or other shorts. While doing this, I discovered that if the circuit was in a non-working state and I shorted out pin 1 and 14 on the IC, it started working! All 5 of the newly built circuits behaved this way.

I lied. 4 of the 5 new circuits behaved this way. The difference is that I had a LM380N leftover from last year that I used on 1 of the new circuits. This one performs without issue and works like the one I built for myself and mislead me to believe I knew what hell I was doing. The 4 circuits that I built using the newly purchased LM380N ICs have to be briefly shorted out from pin 1 to 14 and they start to work until I hit the guitar strings hard, which I assume would correlate to input level, and then they break. Could this be a difference in production runs of the ICs? The question is how to make the current ICs work, or how can I order ICs that are exactly the same as the ones I received a year ago. I suspect the latter is quite impossible.

I have tried to add several more of the 470k Ohm resistor in series to the + of the input jack, but that didn't seem to do what I had hoped. The circuit still cut out with a hard hit to the strings. It seems that the 10uF cap, C5, should be resetting the IC, but I'm not sure why it's not. I did check several of them by checking resistance one way and then reversing the probes and seeing numbers flash on my MM for a fraction of a second. So, I think the caps are good; its unlikely that 5 in a row would be bad anyway, no?

Well, I think I've written enough of the story here. I've resigned to not being able to give these out as Christmas gifts this year. I know this is a long post and everyone has far better things to do, but if you're board and feel like trying to diagnose a problem without the problem in front of you (which I understand is prohibitively difficult), I would be elated to hear feedback from someone who actually knows what the hell he or she is doing. I'd be more than willing to post measurements of different points in the circuit; I'd just need some direction of what you'd be looking for.

Thanks for your time.
Best Wishes and happy Friday.
 

Attachments

  • headphone amp.jpg
    headphone amp.jpg
    45.3 KB · Views: 160

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
It seems that you are overdriving the amplifier. I suggest you play some gentle lute chamber music.

Alternatively, you could make sure the chip is kept within its rated temperature range by reducing the drive and increasing the heat sink size.

Shorting pin 1 to pin 14 shuts off the bias current to the output stage so reducing the dissipation. Luckily, it should not do any damage.
 

davenn

Moderator
Sep 5, 2009
14,011
Joined
Sep 5, 2009
Messages
14,011
and added to what duke37 said ......

it would probably be much wiser to have the level control on the input rather thanthe output !!

cheers
Dave
 

hrtendrup

Dec 20, 2013
2
Joined
Dec 20, 2013
Messages
2
Thanks for the replies! Sadly, I sold many electric lute so I could buy the parts for this project. It looks like I have some research to do on how to reduce the drive and some research on heatsinking. I would have guessed that more resistors in series connected to the input would have lowered the drive, but that may be wrong. Allow, the datasheet mentions something about a heatsink but I didn't read it thoroughly because the project information from the book mentioned in my original post never mentioned a heat sink. It did say that grounding 3-5,7,10-12 would help with heat dissipation.

Also, the ic wasn't resetting itself when it stops working and something doesn't seem right about that.

Anyway, thanks again to each of you for your time, I'll post updates to this circuit saga.
 
Top