Making a BD139 voltage follower buffer, need a little help

[majkel]

Nice too see another advice for 200 ohm resistor removal. If you keep it, then the transistor makes disputable sense in this circuit. I also agree to putting a resistor between the base and the op-amp out, however if the oscillations are about to occur, they will surely do even with the resisitor.
Another thing - I said remove 1M, you placed 100k. :| Next time I say "stand up" when you are sitting, you will lay down.
And the last thing. I HATE OPA2227's sound. Dark, dull, emotionless, boring, annoying, gimme more adjectives...

 

[~n00beR]

Although the value maybe high, I thought the reason for putting a resistor to ground at the input signal was to prevent DC offset if the pot wiper lifts off the contact surface?

 

[majkel]

Quote:

Originally Posted by ~n00beR /img/forum/go_quote.gif Although the value maybe high, I thought the reason for putting a resistor to ground at the input signal was to prevent DC offset if the pot wiper lifts off the contact surface?

This is true but, would it happen? On the other hand, you'd still hear the unpleasant scratch with the 100k resistor. This resistor makes more sense when we build an AC-coupled input, and then AMB's suggestion about the capacitor between 1k and ground is valid. Otherwise - not.

 

[FallenAngel]

New schematic - still a huge offset, but different value, now it's around 300mV

Going to play around with this thing some more later today

 

[TzeYang]

i'm pretty sure you have to install a cap from the 1K resistor to ground.

, at least the Heed CanAmp does so.

And since you're using FET inputs why bother with 910ohm resistors? Wouldn't it be better to use a higher value so to balanced the input bias?

 

[~n00beR]

As i said before the heed amp has a non polar cap in series witht the input signal.
I have found that this cap has the most effect on offset.

Just to test it you could use a polarised cap, and changed to a non polarised cap when you get one.

Using the input cap should take you into the 10's of milivotlts range (30 ish), using the input cap and the feedback cap should take it even lower.

Input caps are not all that bad, many people use them as upgrades to "smooth out" the sound.

Hope this helps

n00b

 

[NelsonVandal]

Now it's almost a Tori amp. Why not use the exact values suggested by PRR. It's supposed to be working. Is your opamp healthy? What does the ground look like?

 

[majkel]

Quote:

Originally Posted by ~n00beR /img/forum/go_quote.gif As i said before the heed amp has a non polar cap in series witht the input signal.

And this is the only case that both caps make sense - .47uF~1uF non-polar instead of 910 ohm resistor, and 10uF in series with 1k. Then the pot can also be of higher resistance with no impact on the circuit.

 

[FallenAngel]

Hey guys,

Thanks for the generous amount of help here, and as I feared, the problem was not with the circuit, but with my oversight in building it. I accidentally shorted the Output of the chip to the inverting input during construction (I plan on paper and then transfer to perfboard, so changes in circuits are harder to track). When I cut the 475K resistor for the inner loop, I attached the two points.

After waking up this morning with a vision in my mind of why this doesn't work (yeah, I'm strange that way, I can wake up in the middle of the night and have all sorts of weird realizations about things I was thinking about earlier). Quickly rushing out of bed and cutting that connection, I get a DC offset of 0.0mV, like NONE. As amb said earlier, the output of of the opamp is around 0.6V, but after the transistor, it's 0mV.

I haven't tested that the amp works yet (no actual input/ouput wired), but I'm think it will.

So.... any suggestions on how to make it sound better? I'm going to lower the gain to about 4 to make it reasonable.

EDIT / UPDATE : Well, being the impatient person I am, I wired everything up and it sounds pretty damn good! Nice and clean, no noise at all with 32Ohm Grado SR225 even though the transformer is within 6" of the amp. Thanks a bunch to all who helped out with getting the design up to this point.

I will be rebuilding this amp to make it smaller and look nicer sometime this weekend, if anybody has suggestions on what to do differently, they would be appreciated. Thanks!

 

[FallenAngel]

Quick question guys, how much heat can these BD139-16's handle? The heatsink gets VERY hot, I've had the amp running about 2 hours (and it sounds wonderful), but the heatsink is too hot to touch. I can't get much info from the datasheet, just that the Maximum Junction Voltage is 150V.

 

[MisterX]

That is way to hot.

 

[FallenAngel]

Thanks, very descriptive

It's not exactly too hot to touch, just too hot to keep a finger on for more than 5 seconds without some kind of meditation.

 

[TzeYang]

i had experience with modding the heed can amp. They are in fact operating at temperatures where i cant lay my finger on the heatsinks for more than 5 seconds.

, not to say it's wrong but i would rather switch to a larger heatsink.

 

[FallenAngel]

I've got another somewhat bigger CPU heatsink around, I'll switch to that one and report back if it still gets hot, but with the size of that monster, I'm thinking it'll be fine.

 

[MisterX]

Quote:

Originally Posted by FallenAngel /img/forum/go_quote.gif Thanks, very descriptive It's not exactly too hot to touch, just too hot to keep a finger on for more than 5 seconds without some kind of meditation.

Ironic you mentioned the 5 second rule of thumb because I had actually typed that in my "very descriptive" reply but removed it because I felt it came across as patronizing when I previewed it.
Don't go overboard with the heatsink here cause you don't need to.
(the heatsink for a tread is better suited for this role then a cpu heatsink)