FET Bass Preamp

[gtrmac]

Hello, noob here. I put together a Bass Guitar Preamp from a design by Albert Kreuzer and it's working but I'm looking for ways to improve the operation. I'll probably strip out some of the bells and whistles like the bass cut and possibly the overload LED circuit since they aren't functions that I particularly need. I added a balanced output using a SSM2142 which works OK. I'm also thinking of just making a version with a passive Fender style tone circuit and see how that sounds. Maybe some of you guys here could glance at the circuit and make some suggestions that might improve it.
Thanks in advance.

Here's the schematic.

Anybody think this thing has potential?

 

[Mendelt]

I found the same schematic some time ago. Never did anything with it but it looks ok.

There are some things i don't really understand about this circuit. It needs a + 0 - supply but for the preamp itself it only uses the + and 0. You could easily get a bit more headroom (6 dB?) when you use the - supply line too. All the connections are AC coupled so I don't see a problem with this. And I don't like all the extra ultra-low, low cut and ultra high switches. It's overkill when you allready have the extensive eq-section.

I never really understood Fets so i don't really get how the gates are biassed.

 

[PRR]

> looking for ways to improve the operation.

What needs improving? The balanced output is good. Otherwise it looks like a fine slightly-flavored bass preamp. Nothing screams "Fix Me!"

> And I don't like all the extra ultra-low, low cut and ultra high switches. It's overkill when you allready have the extensive eq-section.

They don't do exactly the same things that the EQ stage does. The corner frequencies are not the same, and the Cuts go down to infinity. And the "value" of any electronics is proportional to the number of knobs and switches: all these $0.50 cap-switches add a lot of "value" for their cost.

> It needs a + 0 - supply but for the preamp itself it only uses the + and 0. You could easily get a bit more headroom (6 dB?) when you use the - supply line too.

Single-ended amplifiers normally work "best" with single-ended supplies. The grid and cathode (or gate and source) want to connect to signal and power ground. If they run down to B-, you inject power supply noise at their inputs.

What more headroom? Assuming +15V supply, this thing can make 2V at 5% THD, 0.2V at 0.5% THD, mostly simple sweet 2nd harmonic. You adjust Sens, Gain, and Volume to keep signal levels good. Input is around 0.1V, output won't need to be more than 1V or 2V, signal is not un-distorted but far below harsh distortion. Or for "flavor", you crank Sens and Gain to make T2 strain, then reduce Volume to proper output for your power amp or PA input.

If you wanted an UN-distorted sound, 0.01%THD, you would have used TL072 or another fancy-amp instead of the simple FET gain stages.

I once put together something like this, with a very clear/harsh sand-state power amp behind it, for a guitarist, and he was very happy. It isn't dead-clean, and you don't want dead-clean on guitar. Fender-bass often is preamped dead-clean: its distortion could muddy the lead-guitar's role, and there is usually plenty of soft distortion in a speaker reproducing bass. In some acts, control of odd-order distortion could be helpful to add more "heavy" in heavy metal. In some jazz and blues bass, some 2nd harmonic could be useful. Which is why I don't like the question "how to improve?" without context. FOr some acts, "improved" would be 2 feet wide and 97 knobs. For another, "improved" might be reducing it to the size of an iPod-nano. Some players like to work super-clean, getting "expression" with fingering; others make music by playing off the "phatt" sound of an imprecise amplifier.

 

[gtrmac]

Thanks for the replys. I've been running a sine wave through this circuit and 'scoping the signal at the stages and one thing that may be a problem is that there is a lot of gain in the first stage which clips the second stage when you turn the gain pot past about 20%. This is OK when I use it with guitar, a little 'crunch' is useful, but for bass it gets 'mushy' sounding a little too fast for my taste.

I put trimpots at the FET source and drain resistor points to see what effect changing the bias points has and this has helped. I can alter the gain at the first stage to clean up the second stage but it still clips very early. I will try it without the bypass cap on the first stage FET source and see how this works too. The 5k 'sens' pot has only a slight effect on the sound. I substituted a 22uf bypass cap which reduced the gain slightly I think too.

The filter switches can also be useful because you can use them with the EQ switched out.

The master volume before the last FET stage acts weird too. It increases the output smoothly until the last 10% of rotation and then there is a big jump in volume. This is annoying and I'd like to improve it's operation. Maybe I should put the pot after the last FET?

 

[Kev]

I may be a little dull today but what is the difference between the Low and the High inputs ?

What is the + voltage for this one ? 9v ... more ?
more could be good lots more ?
but are the op-amps on the same supply ?

Do you know the frequencies for the Mid select ?

lastly
wots with the Ultra High ?

 

[gtrmac]

Low sees one of the 68k resistors, High sees two of them in parallel.

I'm running it at +/-15V. The FETs are using only the + supply.

I'll get back to you on that, I'm not using the Mid select switch, just the simpler version with a fixed midrange at 1k I believe.

The UltraHi bypasses Hi Freq around the Volume pot, like a Bright Switch.

 

[Bauman]

I must say that the Hamptone FET preamp can be used as bass preamp, and it sounds sooooo good for this. :thumb:

Fabio

 

[Kev]

[quote author="gtrmac"]Low sees one of the 68k resistors, High sees two of them in parallel.

The UltraHi bypasses Hi Freq around the Volume pot, like a Bright Switch.[/quote]

funny
I see it now
I don't think I like the phono representation
for some reason when I look at the High input I see a connected signal go straight to ground.
The contactor is the switch not the "contact"

never mind ... stupid Kev

Do we expect the volume pot to kill some high frequency when in the full position ?

Yes Fabio,
I think any simple clean lined gain device with good headroom will generally make for a good bass preamp. The Hamptone or Joe's DI into any of the good Mic-pres also make for good Bass paths.

I think PRR's idea of fitting this into a small pocket sized toy is a great idea.
An old cigarette packet could be funny. Smokin' Joe Bass pre-amp

 

[gtrmac]

Kev-

When the volume pot is full on the preamp clips like mad and sounds very bad. I need to figure out how to get the amp to function within a range that gives you a line level output with a minimum of clipping. It would be much more useful a a bass pre this way.

Any ideas?

 

[bcarso]

[quote author="gtrmac"]Kev-

When the volume pot is full on the preamp clips like mad and sounds very bad. I need to figure out how to get the amp to function within a range that gives you a line level output with a minimum of clipping. It would be much more useful a a bass pre this way.

Any ideas?[/quote]

The lowest-impact way to reduce gain would be to make the drain load resistors smaller (R4 and R9). I'd start with R9 since it has the largest signal. There will be a little interaction with the response of the network following R4 but nothing too serious; ditto a small effect on the low freq rolloff of the three-band EQ circuit.

There are some details of this schematic that don't inspire confidence btw: no d.c. ref for IC2a n.i. input, useless parts like R14/C9. R1/R2 contribute a lot of noise but make the thing pretty bulletproof. But overall just a straightforward, rather colored preamp.

 

[gtrmac]

Thanks for the reply.

I'm still tinkering with the circuit.

So how do you get rid of some of the color? :grin:

 

[bcarso]

[quote author="gtrmac"]Thanks for the reply.

I'm still tinkering with the circuit.


So how do you get rid of some of the color? :grin:[/quote]

It's a good circuit for learning by tinkering.

Well---as PRR says, most of the distortion per se will be second harmonic, so pretty benign and even desirable for some applications.

If you wanted just clean gain, there are many ways to go about it---the op amps mentioned by PRR, or converting some of the pure FET gain stages to more complicated compound stages with local feedback. I'd look at ways to protect the input stage from overload without the big series R's, and also maybe reduce the variable input capacitance effects. But it quickly turns into a different animal.

 

[gkhebert]

gtrmac,

An additional way to reduce the gain is to eliminate C6.

I don't know if you have implemented the tuner output, but IC2a is relying on leakage through the 100nF cap (C18) for its non-inverting-input bias current. This is not good practice, and could result in this output being stuck to one rail or the other. Add a 1M to ground, as on IC2b.

Regards,
Gary

 

[gtrmac]

Thanks for the reply

I'll try those suggestions and see what happens.

 

[gtrmac]

[quote author="bcarso"]

It's a good circuit for learning by tinkering.

Well---as PRR says, most of the distortion per se will be second harmonic, so pretty benign and even desirable for some applications.

If you wanted just clean gain, there are many ways to go about it---the op amps mentioned by PRR, or converting some of the pure FET gain stages to more complicated compound stages with local feedback. I'd look at ways to protect the input stage from overload without the big series R's, and also maybe reduce the variable input capacitance effects. But it quickly turns into a different animal.[/quote]

National Semi Application Note AN-32 has a "Low Cost High Level Preamp and Tone Control Circuit" which I am thinking about trying.

I definately want to check out the Hampton Pre also.

I removed the source bypass caps C2 and C6 and it lowered the gain and reduced the clipping.

Any suggestions are welcome!

 

[lion]

I've also build the Kreuzer preamp. I get reasonable gain with low distortion through Q1-Q2, but the last stage (cathode follower) is clipping very asymmetricaly (bottom of waveform only). How can a dummy like me mod/adjust the last section for best performance?

Erik

 

[bcarso]

[quote author="lion"]I've also build the Kreuzer preamp. I get reasonable gain with low distortion through Q1-Q2, but the last stage (cathode follower) is clipping very asymmetricaly (bottom of waveform only). How can a dummy like me mod/adjust the last section for best performance?

Erik[/quote]

What are you loading it with? That last stage is only good for light loads, and will be challenged with low impedances. Replacing the FET source resistor with a current sink would help a little. Making it into a "White Follower" would help some more.

 

[lion]

At the moment I'm running it as per the schem in the EQ bypass position.
Could you please explain about the 'current sink' and 'white follower'.

I'm planning to insert a passive (Baxendall) tonecontrol. Any suggestions on which filter values to choose for use in the Kreuzer baspreamp?
I'm aware of the ToneStackCalculator, but I'm not sure about the load values. If I'm inserting the Baxendall instead of the active tonecircuit which loads would the Baxendall see?

Last question - how much output voltage would I need to drive a power amp properly? I'm wondering if I need another gain stage to make up for the loss in the passive tonestack.

Erik

 

[lion]

I've found that by lowering the value of R23 (3M3) to 1M - the waveform looks better, with the clipping of the bottoms almost gone. Is this the way to go - or am I only inviting other problems?

Erik

 

[bcarso]

[quote author="lion"]I've found that by lowering the value of R23 (3M3) to 1M - the waveform looks better, with the clipping of the bottoms almost gone. Is this the way to go - or am I only inviting other problems?

Erik[/quote]

If the change in the gate voltage divider puts you in a better region for output swing then I don't see any reason not to do it.

The problem with FETs is that their characteristics are all over the map, regardless of device type. Some manufacturers will select and sort into narrower zones than others, and you pay a premium for this.

A current sink (or source) is just a circuit that has an output terminal current that varies slightly for a change in terminal voltage. Typically they are made out of a transistor, bipolar or FET, and the output current is taken at the collector or drain. I'll try to find a ref. with schematics. EDIT: Fig. 5 of this ref has a two-transistor version shown: http://www.planetanalog.com/printableArticle.jhtml?articleID=60404792

A White follower is a compound stage which has a current sink in the bottom and senses the upper device's current. The latter signal is used to drive the current sink actively so as to tend to compensate for the current delivered to the load. It helps cathode followers a lot, can be applied to FETs, and is less popular in bipolar designs, partly due to the ease with which low output impedance can be achieved other ways.