Just FYI, this is an older version of the preamp. I believe the schematic for the new design is only available if you order the board. If you look at the project page you can quickly notice that the picture of the populated board does not match the schematic. Notable differences are the fact that the populated board shows two dual-opamp ICs while the schematic indicates only one. The output buffer transistor is also nowhere to be seen on the board. It's probably safe to assume that the new op amp is at least used to replace the output buffer and perhaps more.

It's been a while since I've read that thread but without going back and re-reading it all I believe the improvements were as follows:

-increased input impedance and reduce noise by using a separate voltage divider network to generate the bias voltage and then use a higher impedance resistor to reference the non-inverting input to the bias voltage

-additional power supply filtering

There are probably other improvements that's I'm missing, but those are the ones that stand-out in my memory.


Keep in mind that the schematic attached to the first post of that thread is not "the most improved" design that is presented in the thread. Based on discussions within the thread the design was further refined and updated schematics were presented.

I don't want to speak for the OP, but I can answer your question. When you see trimmers like this on the drain of a JFET gain stage they're typically there to adjust the bias point of the stage. The electrical characteristics of JFETS vary greatly from one to the next (even of the same model!) so one "work around" is to provide a trimmer for the source resistor in a "self bias" configuration. The trimmer is adjusted until you see about half supply voltage on the drain pin and the stage is considered "biased". In reality, these trimmers are usually much larger than they need to be and that can make it very difficult to dial in a good bias point (the trimmer is too sensitiive). You see this approach a lot with designs from runoffgroove.com . You will see a lot of heated debate about the use of trimmers in this way. Some people think it's lazy design, others think it provides greater tolerance of variance in JFET parameters. I believe that for home-built DIY stuff all that matters is if the circuit sounds good to you when you're done. Now if we're talking about a production circuit meant for mass production, then the game changes.

I was reading a TI document about single supply op amp circuits and found something interesting. Everything I've read up to this point indicates a biasing voltage of half the supply voltage should be applied to the non-inverting input to properly bias a single supply circuit.

This document, however, seems to indicate that this bias voltage can be applied to the inverting input instead. As a matter of fact, this seems to be touted as a preferred method.  Here's a link to the document: http://instruct1.cit.cornell.edu/courses/bionb440/datasheets/SingleSupply.pdf Pages 5-6 illustrate the type of circuit I'm talking about.

So, is there really anything to be gained by biasing in this way? Any drawbacks?

Hmm.. you're right. I re-ran my simulation here at work (with the same file) and got the behavior you describe. I'll have to check when I get home but I'm quite sure that varying C1 changed the gain when I tested it there. Here at work the gain only responds to changes in the voltage divider network biasing the top JFET.
-Matt

I've been playing around with some spice simulations of a basic mu amp. In my case I simulated the first gain stage of the blackface preamp found here: http://www.redcircuits.com/Page120.htm

My question relates to controlling the gain of these stages. In my simulation it seems C1 from the linked schematic (the cap from the gate to the source on the upper FET) has a huge effect on the gain of the circuit. Is this, in fact, the easiest way to control the gain? My simulations didn't show any serious change in distortion characteristics by just changing the cap because it doesn't affect the bias point of the stage.

Is it really that simple? I know simulations aren't perfect so I'm wondering if there's some real-world interaction that I'm missing.

I've attached a LTSpice file of the circuit in case anyone wants to play around with it.

Not a problem. It happens all the time I should have been more clear which circuit I was commenting on.

As for your circuit, Have you tried removing the first op amp buffer? (immediately after the tone stack). Since it provides no gain I don't see how it would affect the sound of the circuit. Since the final gain stage is non-inverting it already has a high input impedance (Assuming you moved the 2M2 R5 to replace the 47K R6). It would be interesting to move the U1a buffer stage in front of the tone stack to isolate it from the pickups in case a guitar is plugged in directly. This would  help prevent loading of the pickups.  You mentioned that you tried adding gain in front of the tone stack but like this configuration better... did you try just a buffer with unity gain?

-Matt

I should probably clarify, I was talking about the circuit posted by James when he started this thread, not the circuit you posted as a reply. The input stage of that EQ circuit (the one posted by James) is an inverted op amp gain stage. I'm not quite sure what you're talking about things being more complicated than that.

From here:
http://www.electronics-radio.com/articles/analogue_circuits/operational-amplifier-op-amp/inverting-amplifier.php

"It is very simple to determine the input impedance of an inverting operational amplifier circuit. It is simply the value of the input resistor R1. This is because the inverting input is at earth potential (i.e. a virtual earth) and this means that the resistor is connected between the input and earth."

Which I believe is what I said. The circuit James first posted has a series resistor between the signal source and the inverting input of the op amp. I'm not sure what other resistor is in parallel with it if you assume the inverting input to be at ground.

Now, if we were talking about the non-inverting input then you're quite correct. You have to take into account the resistance of the bias reference resistor (assuming a single supply is being used) in parallel with the internal resistance of the op amp. Now while its true that the internal resistance can vary greatly between types of op amps most modern JFET-input devices (like the TL07x series) have very high input impedances. According to the data sheet the internal impedance of the TL072 is 10^12 Ohms which is 10 TERRA ohms. That huge value in parallel with any resistor in the 1M - 3M range will have pretty much no impact.

I also wouldn't suggest that circuit as a preamp by itself. While it does have adjustable front-end gain it isn't much. At the low end of the adjustment it's actually attenuating the signal (gain of 0.10x so your output signal would be 1/10th of the input) at max gain the amplification is only 2.6x. This can be easily changed  by altering the component values though. The bigger issue is the input impedance. Having an inverting op-amp as the input stage means that your input impedance is dictated by the inverting input resistor. That means an impedance of only 100K. You want at least a 1M input impedance if you're directly plugging your guitar into this circuit.

As I'm sure you're aware this type of circuit will not be good for getting any distortion sounds. So if you're just after a clean preamp then why not just roll your own circuit? If it were me I would just use a non-inverted input buffer stage with a 1M input impedance. Then follow that with an active baxandall tone control. You could pull the entire thing off with one dual opamp and a handfull of resistors, capacitors and pots. You make all sorts of variations on this design. Change the tone control to a passive FMV and add a recovery stage, add clipping diodes in the gain stage(s), etc...

Edit: Here's an example of what I'm talking about.
http://www.runoffgroove.com/tonemender.html

That has a non-inverting input buffer with unity gain (easily converted to have positive gain), a passive FMV tone stack with some adjustments to shift the response around and then a recovery stage with additional gain.

If it were me I would probably convert the input buffer to a gain stage with adjustable gain and change the recovery gain stage to be fixed gain close to the average insertion loss of the tone stack (maybe 10 - 15dB or so) then just add a master volume control after the whole shebang. I believe it's better to do your amplification up front rather than try to recover too much volume from a depleted signal. Having your gain up front where the signal is strongest and most "clean" (from the guitar) should reduce noise.

Can you elaborate on what you mean by the trides not "having what it takes"? To do what? I'm under the impression that if an amp sounds good driving a speaker it will produce a similar sound (give or take) when driving a dummy load and "tapping" the output with a following stage. Using a reactive speaker motor go even farther toward replicating the signal driving a speaker.

regulators work very well as long as your power amp supply voltage falls within the limits of the chip. Most of the standard 78XX series have a maximum input voltage of 35V. If you need to go higher than this you may need a custom "regulator" built from a zener diode, resistor and capacitor. Take a look at the power supply section of the preamp on this page: http://sound.westhost.com/project27.htm Note: You will need different parts for a 9 or 12V pedal and only half as many rails (that preamp uses a bipolar supply).

Do you have any sound clips? Also, what kind of power amp are you using?
-Matt

I just finished the first stage of my hybrid combo amp. My plan was to build a complete low-wattage tube amp, run it through a dummy load (probably a weber MASS speaker motor) and tap the speaker level to drive a passive tone stack followed by a DSP effects processor and finally into my LM3886 power amp. The combo will likely use a single 12" Lil'Texas neo speaker to keep the overall weight down.

I have completed the first phase of my master plan The tube amp is done and tested. It's a firefly from www.ax84.com
If anyone is curious I have a 2mb sound clip of my friend noodling around on the amp. Keep in mind that the amp was only driving a 5" Jensen MOD speaker so the bottom end is kinda muddy. Strange, who knew a 5" cone didn't do so well in the low range  . The guitar is a Gibson ES 335 with stock humbuckers.

http://www.mr2-power.com/firefly%20test%201.mp3

-Matt

Sounds great for really high gain stuff... I know you already answered the question about clean sounds, but do you have any samples? I'm curious how well it cleans up. Congrats on a great build

Welcome! I'm pretty new here myself but thought I would try to answer your question. Things move kind of slow on this board compared to many others I'm on (not a huge amount of activity) but the quality of the information is generally very high. There's a great group of regulars that have answers to nearly any question you could have about SS amps (and amps in general!). You just have to be patient for replies to arrive

As for where to begin... while I'm mostly a beginner myself (only a couple of builds under my belt) I would suggest investing a lot of time researching preamps. There are some great threads in the preamp section that can provide hours of reading. While the power amp section is important in SS amps it is pretty widely accepted that the preamp is most responsible for the type of tone you get out of your amp. The SS power amps are generally designed to be very "transparent" and do not color the sound too much. So, research preamps... Listen to clips when you can find them and try to decide what you want your amp to sound like. I wouldn't try to reinvent the wheel your first time out so just choose a nice simple preamp circuit to build. On good route to take is to choose a DIY effects pedal that you like ( www.runoffgroove.com has a number of them) and build it. Once you have built and debugged the pedal to your satisfaction then you can typically use it to drive your power amp. While a bit more expensive, a good option is to buy a pedal kit from somewhere like www.olcircuits.com or www.generalguitargadgets.com They will include everything you need and give you the best chance of success.

For the power amp I would suggest going with one of the "chip amp" designs. While there's a lot to be said for sitting down and designing a discrete power amp from scratch the reality is that an IC based chip amp will have fewer components and be easier to build and debug. As with any hobby, people stick with it longer and have more fun if they have good initial success.. If your first build fails and is riddled with problems then you're not as likely to continue

So, if you assume you're going with a chip amp kit then the choice of power amp will come down to how much volume you need/want. The LM3886 kits sold by Brian at chipamp.com are very well regarded and can be VERY loud (40-68W depending on how you build it and what impedance your speaker is). Brian also sells a LM1875 chip amp that is lower power but still should be loud enough to gig with. I've only built the LM3886 kit but it was a very straight-forward build with great directions.

If you want something low power and easy to build I would suggest TDA2005 based kit found here: http://store.qkits.com/moreinfo.cfm/FK607 As far as amp kits go it's as cheap as they come ($10 for the lot!) and it has the advantage of not requiring a fancy power supply. You can simply power it with a 12V 2A DC power supply found anywhere. This will avoid you having to deal with mains voltages your first time out which can be intimidating if you're not used to working with that kind of voltage. You may be able to find a supplier for futurekit projects somewhere in the EU to save on shipping costs. http://www.futurekit.com/distributor.php

After you've built and testing your preamp and power amp combination you need to start thinking about an enclosure. This is where things get interesting (for me, at least) you can do a head only (probably the easiest) or a combo amp... Many people buy cheap/old amps (working or not) and gut them to avoid having to do lots of woodwork. That's all up to you.

Hopefully this will give you somewhere to start... also I'm hoping someone else will reply as well to give you another point of view.