Serious Blocking Distortion

[Littlewyan]

Tried the amp tonight and found that it still has too much bass, so I'll try what you said Roly and lower the 22uF Cap on the JFET Source to 1uF.

Also I found it overheats on full volume pretty easily! So I've just ordered a much bigger heatsink for the Voltage Regulator as the current one has a rating of 21C/W, compared to the heatsink I just ordered which is 12C/W. When the amp is at full volume the regulator will be dissipating 6.65watts, so multiply that by 21 and you get 139 Degrees C, which is above the regulators 125 Degrees limit. With the new heatsink it should run at a maximum of 79.8 Degrees C.

Other than that it sounds good, even my rhythm guitarist liked it and hes very picky!

[Littlewyan]

Also Roly what is that simulation program you're using?

[Roly]

LTSpice, available for free download from Linear Technology, no strings, Getting Started pdf, fairly compatible with other flavours of Spice.  Also Yahoo LTSpice group stuffed with models and an active forum.

Murphy's Law of heatsinks - they are never big enough (you always need more than you thought).

[Littlewyan]

If I change the resistor values for the BJT to ones you have specified won't that lower the output of the BJT?

[Littlewyan]

Ignore the last question, downloaded and ran LTSpice, seems that I get increased output due to the lower current consumption of the BJT. However I found that after adopting the values for the BJT that you specified it didn't actually seem to do much at all! Audio Output was the same before and after the BJT. So what will it actually be doing if I adopt the values you specified?

[Roly]

The BJT stage is drawing much more current than it needs to, and in conjunction with the decoupling resistor pulling the supply rail down considerably.  By simply scaling the resistors up and current down the supply rail goes higher giving more headroom and incidentally biasing the FET stage better.

[Littlewyan]

But the audio output from that BJT is the same as it's input if i use those values so what is it doing? With my values it increases the audio signal but with yours it stays the same. I understand the reason for using your values but what does the BJT do?

[Littlewyan]

Also the overheating issue is now (hopefully) fixed. I have fitted a much better heatsink which has a rating of 12C/W and the difference even at low volumes is extreme. So I shall give the amp another thrashing at band practice tomorrow night to see if it will hold up. I must say the chip I'm using for my output stage runs very cool considering its pushing out quite a bit of volume!

I shall now modify the resistor values to Roly's spec.

[Littlewyan]

Nearly there guys, here is an updated schematic for the amp so far. I'm pretty happy with it apart from there is still a tad too much bass, so I'm going to replace the JFET Source Cap with a 0.68uF Cap later in the week if I can.

[Roly]

what does the BJT do?

Currently, not a lot; it's just lurking until you decide you want to give the diode clipper another go I guess.

[Littlewyan]

O well i'll just leave it in there. Tested the amp at band practice, plenty of volume, still bit too much bass so i'm going to try a 0.68uF cap on the JFET source.

[J M Fahey]

Try this:

[Littlewyan]

Thanks for that J M. Would .68uF be too much on the Emitter of the BJT? As I don't have a cap as low as .1uF.

[Littlewyan]

Here is the amp now, I have made some of the changes you specified J M, simply because I just tried the additional resistor and capacitor to begin with and that made a world of difference. The amp is so much better now. I need to test it at band practice really though to hear it at a good volume. My only possible quarrel with it is the overdrive is quite harsh but as I understand it (correct me if I'm wrong) this is a characteristic of transistors. I know Marshall have made a very good Transistor amp that has nice soft valve clipping but their amps are pretty complicated compared to this one! Is there any way to soften the overdrive?

[stormbringer]

In my opinion you can get quite warm distortion from jfets, i would suggest trying different types, and maybe make the gain adjustable.

There really isn't any magic about distortion, filtering Before and after to shape what frequencies get the majority of the distortion etc.

let's take a well known tube amp (analysis is from ltspice, not sure how accurate):

The first stage is a Clean amplifier stage that increases the signal level and rolls of frequencies below 70 hz.
the second stage clips the + side of the signal, filter after rolls of some of the highs
the third stage clips the - side of the signal, filter after increase high mids
fourth stage clips the entire signal, and drives the tone stack. which schoops out the mids and goes to a buffer stage.

This is a high gain amp, but as you can see, there is alot of filtering going on, not just bringing the signal up to clipping.

But also, as i said: try a different jfet aswell, mpf102 and 2n5457 are quite easy to obtain and used in lots of guitar effects, you will have to rebias them though.