Transistor bias, changing PSU voltage on compressor circuit ?

  • Thread starter Thread starter Blue Jinn
  • Start date Start date
Blue Jinn

Blue Jinn

Rider of the ARPocalypse
This is kind of a followup to my other post on line level stomp boxes. I still want to run some of these off an existing (for another circuit) +/-15v

In the attached, the IC connection is pretty straight forward. As to the transistors, I'm not quite sure how to approach that as to the bias for Q1 and Q2 (or if that would necessarily matter going from 9v to 15v. I have an app called AppCAD, which will calculate all the resistor values for that circuit fragment. I don't exactly know how to use it, but it does seem pretty straightforward, plug in the transistor paramters (Hfe etc) and then plug in the voltage range you want at certiain pointes and then some baffling temperature coefficient from emitter to base. THen you plgu in Vcc, Vce and Ic. The AppCAD shows you where all these values occur. In the schematic these are different for Q1 and Q2. I am assuming this is because you want different gain at both those two places, but I have no idea how to figure this part out. If is a matter of gain, I am not exactly sure what is needed where. I am also assuming that Q2 drives the side chain.

Any guidance is helpful appreciated.

The schematic can be found here:

tonepad -- file information
 
Hey Blue.
How's it going?
Have you tried simulating the circuit in LTSpice.
 
Can't see a direct link to a schematic and I am not chancing this new build on picking up useless or damaging crap and I refuse to go through yet ANOTHER ballaching registration!

If a simple common emitter circuit with "H" biasing is required the design process was worked out long before apps!

Chose Ic then Ohm's Law gets you collector load where Vc is 1/2 supply (30V) . Chose Re to drop about a volt then add Vbe and again with OL calculate a bias chain pulling about 5X likely Ib.

The gain of a single transistor stage is indeterminate but ball park? 50dB. It depends a great deal on the source and load impedances.

But, if you just want "some gain" why not an opamp? A TL0 or NE5X will be little more expensive than a BC184L and its additional components.

Dave.
 
As an alternative, could you not just regulate the 15V down to 9V with a 7809 or something?
 
Dave,

You can get the schematic, but you have to view the related project page first.

Click on Blue Jinn's link above.

Click the link ibanez_cp9 (just above the download button)

Then go back and click the download button and you should be able to access the schematic.
 
PRHunt said:
Dave,

You can get the schematic, but you have to view the related project page first.

Click on Blue Jinn's link above.

Click the link ibanez_cp9 (just above the download button)

Then go back and click the download button and you should be able to access the schematic.
Click to expand...

Nah, can't be doing with all that. They want the help? Post an attachment.

Dave.
 
Thanks Dave

Goal here is to rack the attached, using teh prefab PCB and existing bipolar supply.

I didn't realize the link was such a rabbit hole.

The circuit fragments I'm looking at are the bias for Q1 & Q2.

Chose Ic then Ohm's Law gets you collector load where Vc is 1/2 supply (30V) . Chose Re to drop about a volt then add Vbe and again with OL calculate a bias chain pulling about 5X likely Ib.
Click to expand...

I think I see what you are saying and AppCAD seems to calculate values along the lines of the formulas here ---warning another link! -- http://www.electronics-tutorials.ws/amplifier/amp_2.html but none of those two circuit fragments at Q1 and Q2 seems to follow exactly any of the biasing methods I've come across, that is what I'm not really understanding, nor where selecting the starting point. I am assuming that the Q1 is a gain stage for the signal itself and that Q2 is an amplifier for the side chain. I *think* I should be able to extrapolate the current from the present values, I=R/V and go from there, and use AppCAD rather than a pencil and calculator! , but again none of the bias networks in AppCAD or at the site above seem to fit. I'm missing something I know, or interpreting the schematic wrong.

I'm also assuming that IC1 isn't going to care either way what it's getting via Q5, pin 13 (Vref right?) and the side chain so I can leave the rest alone. Or maybe the whole thing doesn't care about the supply voltage, soemthing just tells me I need to do some adjustments though.

It seems somewhat easier to change resisitor values on an existing PCB (which I want to go with here) instead of adding an additional 9v supply. I'm also under the illusion that I'll get better headroom with +/- 15v
 

Attachments

Last edited:
HTTP 301 This page has been moved

That ^ is the data sheet for the IC and you can see the bare bones of your circuit in the VCA diagram.

The chip is happy to have a dual 18V supply but I am not so sure about the rest of the transistors? Yes a 30V supply would greatly increase headroom IF all else scaled the same, but in complex, DC coupled circuits it hardly ever does!

I would take a totally empirical approach. With a 9V supply set all the pots to their mechanical mid point and then derive a voltage table of all the transistor CBEs and the pins on the chip.
Then put say 15V instead of 9 into the beast and check volts again. Have they simply scaled? (good) or gone a bit awry ? (bad) . If things look ok go for broke and 30volts. I would not bother with a split supply,'s'all done for you to get the right DC conditions.

BTW, assuming you want the compressor for "studio" applications (hence the headroom)? I would replace Q1 with a TL072 (or a better fet input chip) to give an "always buffered" input and the other 1/2 on the output to buffer that bad practice 100k VC!

Dave.
 
You must log in or register to reply here.
Back
Top