ruby circuit is "too clean"

[armstrom]

So I built a ruby circuit (slightly modified) to use as a preamp to my TDA2005 power amp and while it has good tone and is quite loud I can't seem to get the amp to "break up" or create any overdriven distortion even with the volume and gain cranked.

Here are the differences between what I built and the ruby schematic:

-removed the filter/zobel network on the output formed by R4 and C7 (not really needed since I'm not driving a speaker directly)
-reduced output cap to 1uF (again, not driving a low impedance load so such a huge cap isn't needed)
-Added a "standard" Big Muff tone control between the preamp (ruby) and power amp.
-Running at 12V
-Added the "grit switch" mod from the noisy cricket

Other than that, it's a standard ruby/cricket circuit. My initial tests were done without a grit switch. Due to the frustratingly clean output (loud as hell, but clean) I decided to add the grit switch and have my buddy try it again... I'm still awaiting results but I'm not too hopeful. I'm using a National LM386-4 and have tried a number of MPF102 fets in case I'm getting too much attenuation from the buffer.. no dice.

Does anyone have any idea what could be keeping the circuit clean? Could it be the 12V giving lots of clean headroom? Or the fact that the amp isn't driving a low impedance load?

I would like to be able to achieve results like this: http://www.youtube.com/watch?v=vk6nd3YcKDc

I want to run the TDA2005 circuit at 12V minimum to maintain good output to the speaker but may try a voltage divider or regulator to get 9v to run the ruby preamp and see if that helps give some overdrive/clipping sound.  Any other suggestions?

Thanks in advance,
-Matt

[brett]

Hi
RE: Added a "standard" Big Muff tone control between the preamp (ruby) and power amp.

There's a big loss in these, especially if the next stage has low input resistance (should be at least 100 kohms).  Try using the BMP output (buffer and gain) stage as well as the tone stack.
cheers

[armstrom]

Hi
RE: Added a "standard" Big Muff tone control between the preamp (ruby) and power amp.

There's a big loss in these, especially if the next stage has low input resistance (should be at least 100 kohms).  Try using the BMP output (buffer and gain) stage as well as the tone stack.
cheers

I thught all the clipping and "distortion" came from clipping the LM386 chip. If that's the case then it doesn't matter what kind of circuit I put after the output as the distortion will have already occured. I may be reducing the overall volume level, but not the other characteristics of the sounds. This is similar to nearly every other distortion pedal out there with a volume pot as the last component in the signal path.

[Steben]

True enough.
Maybe check pins 1,5 and 8 again and reread the gain settings?.

[B Tremblay]

Hi
RE: Added a "standard" Big Muff tone control between the preamp (ruby) and power amp.

There's a big loss in these, especially if the next stage has low input resistance (should be at least 100 kohms).  Try using the BMP output (buffer and gain) stage as well as the tone stack.
cheers

I thught all the clipping and "distortion" came from clipping the LM386 chip. If that's the case then it doesn't matter what kind of circuit I put after the output as the distortion will have already occured. I may be reducing the overall volume level, but not the other characteristics of the sounds. This is similar to nearly every other distortion pedal out there with a volume pot as the last component in the signal path.

Brett is right.  The loss caused by the tonestack is reducing the amount of signal hitting the 386, which is then not overdriven.

Also, I'd suggest using a preamp circuit for the preamp for your 2005 rather than the Ruby.

If you want to use the Ruby as your preamp, then you should have an input gain stage, the tonestack, then the stock Ruby.

[Steben]

Brett is right.  The loss caused by the tonestack is reducing the amount of signal hitting the 386, which is then not overdriven.
If you want to use the Ruby as your preamp, then you should have an input gain stage, the tonestack, then the stock Ruby.

What is this then?
http://runoffgroove.com/grace.html
The tonestack comes between the 386 and the power amp! the 386 is not the power amp.
If the 386 chip is designed to drive down to 8 ohms, why can't it drive a feeble tonestack? 

[earthtonesaudio]

It'd probably be easier to give advice if you could draw a schematic of the whole preamp section and at least the input section of the power amp.

That said, I would suggest first to follow the general debugging steps, then my next thought would be to add back in the resistor/cap to ground at the output.  It's possible the BMP tone stack isn't letting the output swing as far as it can.

[ItZaLLgOOd]

True enough.
Maybe check pins 1,5 and 8 again and reread the gain settings?.

I'm with Steben.  Can you change the volume with the gain pot?  Did you do the Bassman mod that puts a fixed resistor where the gain pot goes?  That would limit the distortion to almost none.

[armstrom]

To address some of the questions:

-The tone stack is after the LM386 chip so it cannot affect the signal reaching the chip.
-I can draw up a schematic if you like but it's basically a standard ruby with the changes I specified in my first post.
-After reviewing the datasheet for the LM386 it seems clear to me now that the combination of increased output impedance (lower current) and a higher supply voltage is making the amp so clean. By dropping the voltage down to ~7V I should be able to get the same breakup as a 9V supply driving an 8Ohm speaker (according to the peak-to-peak voltage plot in the datasheet)

Here are some MP3 samples (about 2mb each) of the amp with the 12V power supply. The first clip is the amp itself. As my earlier posts state it's a ruby preamp, big muff tone control and TDA2005 power amp. The speaker is a 6" Weber signature ceramic.


Note: The amp is called "The Hurricane" because it's a gift for a blues harp player nicknamed "The Hurricane". In keeping with the hurricane theme I called the gain control "category" so when you hear that term that's what is being mentioned (it goes from 1-5, of course ). The "toggle" that is mentioned is the standard grit switch mod from dano's noisy cricket. That will ultimately be labeled "Storm Surge".

Sorry the first clip is so quiet... the amp is quite loud even with the volume only half way up so the narration is hard to hear. I couldn't increase the volume any more without clipping the WAV. The second clip through the PA is a bit better.

Amp Test:
http://www.mr2-power.com/hurricane%20test%201.mp3

Amp through Bose PA system (Bypassing power amp and internal speaker):
http://www.mr2-power.com/hurricane%20through%20PA.mp3

Once I get the amp home and can take some pictures of it I'll make a new thread with the pictures and clips together. I'll also post new clips once I reduce the voltage to the preamp and see if I can get some dirt out of it


-Matt

[B Tremblay]

Brett is right.  The loss caused by the tonestack is reducing the amount of signal hitting the 386, which is then not overdriven.
If you want to use the Ruby as your preamp, then you should have an input gain stage, the tonestack, then the stock Ruby.

What is this then?
http://runoffgroove.com/grace.html
The tonestack comes between the 386 and the power amp! the 386 is not the power amp.
If the 386 chip is designed to drive down to 8 ohms, why can't it drive a feeble tonestack? 

Well, there's no tonestack in either Grace or Big Daddy, so I don't understand why you referenced them.

Apologies for my mistake in reading the original post.

[armstrom]

Well, dropping the voltage did the trick. I dropped the supply voltage to the ruby preamp circuit with a simple 50% voltage divider (two 150 Ohm resistors). The results are nice dirty, bluesy distortion (IMHO). Here's a clip:
http://www.mr2-power.com/Hurricane_test_2.mp3

Any feedback/comments/criticisms welcomed and greatly appreciated!

-Matt