Hey hey! Long time lurker, first time poster, but I’m having some challenges with an output mixer that I’ve built based on the Moritz Klein / Erica Synths EDU Output Mixer (this one: Erica Synths - EDU DIY Output ).
I’ve modified the schematic to have 4 inputs instead of 2 and also wanted to try out some PCBA so there are some surface mount components in my layout of the circuit (though there are still a lot of through hole components (including the BD139 / BD140 transistors)). Having assembled everything it works as expected and I get output and panning and levels working correctly on all 4 channels.
The problem is that the output transistors are getting really hot. Having had it on for 10 mins+ they get up to 90 C (though they may get hotter - I unplugged after that). The original manual for the output mixer suggests that the output transistors are not being driven to the max so they do not have a heatsink suggested. I don’t think that my modifications to add extra channels would increase the load on the transistors, so wondering what might be happening here.
Should I try some different versions of the same transistors in case these are faulty ones (though they do seem to be functional)? Or should I get some heatsinks to just deal with the heat that they produce? Or is there something else that I should try?
Super grateful for any and all ideas. Schematic attached below. Thanks!
I think that R29 and R58 in your schematic correspond to R33 and R34 in the mki x es.edu Output Mixer. I’m not 100% sure but DNM* might mean “do not mount” so I’d give it a shot removing those from your circuit.
Hey! Thanks for the quick reply! I saw a discussion about that on Modwiggler (here: mki x es.EDU Output Mixer - Page 2 - MOD WIGGLER ) and it sounded like they were just to fine tune / tweak the levels for the line out and could be there at a value that works best, so hoping that it’s not them causing this issue…
Is this with headphones plugged in? In that case the load is about 49 Ω plus the headphones’ impedance. Maybe a total of 100 Ω, and then if the signal is up near 12 V the power is 144/100 = 1.4 W. Realistically the average signal would be much less, so that should still be well within the 12.5 W a BD139 can handle — though per the datasheet it’ll heat up if above 1.25 W, but even that’s higher than what you should be expecting of it, so I’m surprised they’re getting so hot. Even if your phones are super low impedance it seems strange.
If headphones aren’t plugged in then the load they’re seeing is much larger (at least 1k) and that really shouldn’t be a problem. If it’s heating up without phones plugged in there’s definitely something wrong.
Oh. Yes. Definitely something wrong.
Added:
If the quick simulation I’ve done is about right (and if the BD139/BD140 models I’ve downloaded are good) it looks as though the collector current should never be more than about 10 mA.
Thanks for the super detailed response. Then maybe I’ll try disconnecting one of the collectors and measure the current going through it to see if there’s something awry there.
An update: swapped out the bd139/bd140 pairs for ones from a more reputable source, strapped on some small heatsinks and now there doesn’t seem to be really any heat kicked out of them at all (and everything works as before).
I imagine the issue was the previous transistors, and now the new ones have heatsinks too, so problem solved! Thanks for all your help!
