G&L Bass Preamp 1.3 frequency plots n fun

[Paul Gagon]

OK, I thought this might be fun.

I've been experimenting with the G&L Bass Preamp (revision 1.3) and wanted to see what kind of modifications made sense.
What I have here is a series of Audio Precision plots showing the effect of component value changes. The PCB used for these
plots has the following components. LM4250 chip, 1% metal film resistors, 50v electrolytic caps and 100v poly caps.

Before we begin, here's the schematic of the bass preamp circuit. Revision 1.3

The above schematic will be used as a reference for the various plots. The component values shown here will be referred to
as the default values. The plots that follow will be showing the effects of altering the default component values to adjust
frequency response, gain levels and filter points of the treble boost.

But first, a picture of the PCB under test.

Note: This PCB is manufactured by Cort and is used in the G&L Tribute line of basses. The Cort PCB normally uses
5% carbon film resistors, lower voltage capacitors and the TL061 chip. However, I modified this PCB to simulate
the USA preamp, which uses 1% metal film resistors, higher voltage capacitors and the LM4250 chip.

Now, on with the plots.

Standard Preamp rev 1.3 as found in the current G&L L-Series basses

RED TRACE = Preamp Bypasses
BROWN TRACE = Preamp On
BLUE TRACE = Preamp On w/Treble Boost

Preamp plots adjusting the value of R7 (27.4K)

RED TRACE = Preamp Bypassed
MAGENTA TRACE = Preamp On w/R7 value at 121K
BROWN TRACE = Preamp On w/R7 value at 39.2K
BLUE TRACE = Preamp On w/R7 value at 27.4K (Default Value)
GREEN TRACE = Preamp On w/Treble Boost and R7 at 27.4K

Preamp plots adjusting the value of C3 (.01uf)

RED TRACE = Preamp Bypassed
BLUE TRACE = Preamp On
BROWN TRACE = C3 value changed from .01uf (Default Value) to .1uf

Preamp Plots adjusting the value of R5 (21.5K)

BROWN TRACE = Preamp On w/R5 value at 0 ohms (jumper wire)
BLUE TRACE = Preamp On w/R5 value at 21.5K (Default Value)
RED TRACE = Preamp On w/R5 value at 68.1K
MAGENTA TRACE = Preamp On w/R5 value at 100K

Preamp Plots adjusting the value of C6 (.015uf)

RED TRACE = Preamp Bypassed
GREEN TRACE = Preamp On
BROWN TRACE = Preamp On w/Treble Boost. C6 value at .015uf (Default Value)
BLUE TRACE = Preamp On w/Treble Boost. C6 value at .022uf

OK, I know this is a whole lotta crazy but I just thought people would want to know what the options are for modifying
this preamp. My hope is that this might get some interesting information out there and see if others find it interesting.
This is a very simple preamp but there's a lot of adjustments that can be made to tailor it to an individuals need.

Anyway, Enjoy
Paul

[TDR1138]

Cool thread. By doing this, do you question the stock values at all? Or rather, what changes would you make?

Side note - not sure if you've seen this this thread viewtopic.php?f=9&t=787, but we were monkeying around with similar ideas.

[Paul Gagon]

Thanks TDR1138,

I was not familiar with that thread you mentioned but after checking it out I gotta say it's awesome. I really enjoy finding some kindred spirits here. Thank you for bringing that to my attention. As for my preferences on component values I think I would have to go with the following.

R7 would change to 39.2K (This gives a broader perceived frequency response)
R4 would change to a 1meg (This along with the R7 change reduces the overall gain on the preamp)
R10 would change to 464K ( This reduces the noise level, increases the slew rate, increases the output headroom and
only bumps the chip current consumption up to 128ua)

Now, this is only my opinion on these changes. I would not want to have people get upset with me changing Leo's preamp sound. I have just found that there is quite a bit of flexibility with this circuit and there might be some players that enjoy experimenting with the tones. Oh! Something interesting about R10. This is connected to pin 8 on the IC. Of the IC chips that are normally used on the preamp (TL061, LF441, LM4250) only the LM4250 has a chip hookup to this pin. The other chips do not have connections to pin 8 and therefore are not using R10. R10 is used on the LM4250 to set its input bias current. This, in turn, sets the operational current of the device as well as the noise spec, slew rate, bandwidth, headroom and some less interesting stuff.

I measured the current draw of the chips I use around here and here's what I got.
TL061 runs at 175ua
LF441 runs at 172ua
LM4250 runs at 50ua with R10 at 1.5M, 95ua with R10 at 681K, and 125ua with R10 at 464K

Sorry, I just realized I'm rambling. Anyway, thanks for taking interest in this. It really is fun isn't it?

[Ken Baker]

Sorry, I just realized I'm rambling. Anyway, thanks for taking interest in this. It really is fun isn't it?

That was a really excellent ramble!

Ramble On!

Ken...

[Paul Gagon]

There is an excellent post that Ken did on upgrading a revision 1.3 Cort PCB to revision 1.3 USA PCB. I made some comments on what he posted and I think I should add them here as well. Here's what I added.
---------------------------------------------------------------------------

Wow Ken,

This is excellent. I've been playing with the Cort board for a few days now. I thought I would share some of my more interesting findings. First up, as you have stated many times and in several posts, the sound of all the various versions are essentially the same. The real issue is circuit noise (caused by using Carbon Film vs Metal Film resistors and the TL061 chip). In the case of the Cort circuit boards you have the additional issue of unit to unit consistency due to the 5% vs 1% resistors.

I started with a Cort circuit board (exactly like the picture you've shown) and did a series of tests on my Audio Precision. The most interesting measurement was the noise level. With the board in its original state I got a noise level of -95.8dBu.
By simply changing all the carbon film resistor to metal film the noise level dropped to -97dBu. When I replaced the TL061 chip with either the LM4250 or the LF441 the noise level dropped all the way down to -100dBu. That's a total noise reduction of 4.2dBu by simply changing to metal film resistors and using a better chip.

Let me address the upgrading to 1% Metal Film resistors. The frequency filters of the preamp are set by the resistors and capacitors on the board. If you have 5% tolerances on your resistors vs 1%......well......guess how much tolerance you're going to have in your filters. As for the capacitors, the ones on the Cort board aren't all that bad, they just could be better. I'm still playing with this circuit and will update you with more fun stuff as I get it.

Oh, I found this interesting as well. This has to do with the power requirements of the various chips. Both the TL061 and the LF441 run at around 175 micro amps. The LM4250 runs at 50 micro amps (with the 1.5M resistor, R10). I can get an additional noise level drop of 1.2dBu on the LM4250 if I lower the resistance value of R10 to 681K. The power requirement goes from 50 micro amps to 95 micro amps but it's still nearly half that of the TL061. Bottom line here? Battery life is longer with the LM4250 over the TL061.

Sooooo......what's my final thoughts. The circuit boards that use the LM4250 chip, 1% Metal Film resistors and good quality caps is a much better unit and something I would want in my bass......well actually.....they are in my bass. Anyway, I'll keep you posted on how things go.
-----------------------------------------

Ken was also kind enough to send me a chip that he was trying. It's the OPA704 chip. This is a CMOS Op-Amp with the ability to have its output signal swing reach within milivolts of its supply voltage. What does that mean to us? More signal level before clipping. It's noise level is just a bit higher than the TL061 but it has much better headroom so if noise is really not a concern for you, the OPA704 is an additional option for the bass preamp. Oh, the power consumption of this chip is right in line with the TL061 and the LF441 (175ua). I have been told that this device is slated for obsolescence so if it sounds interesting, move quickly.

[Paul Gagon]

I also have a description of the various revisions that have been made to the G&L Bass Preamp through the years. I have posted this in two other threads but I think it might also belong here as well. The original preamp that G&L was using, before BBE took over in 1991, used an LM4250 chip and very large resistor values around the circuit. These resistors were of the carbon film type and were typically 5% or 10% tolerance. I was told that there were some complaints about the preamps being noisy (circuit noise, not 60Hz hum. Think hiss vs hum). This was primarily caused by the large resistance values coupled with the type of resistors used.

Rev 1.0 was the first preamp to address the noise issue. It was done in September of 1992 and used the LF441 chip and value scaled down the resistors. These resistors were made from metal film and had a tolerance of 1%. The frequency response was the same as the original but the circuit noise was quite a bit less. I took great care in evaluating the sound of this preamp against the original with a ton of bass players. No way was I going to mess with the sound of the bass. I just wanted to reduce the circuit hiss.

Rev 1.1 was done to go back to the LM4250 chip because it was being so difficult to get the LF441. Besides, with the proper value of the bias resistor that connects to pin 8 of the 4250 you can achieve the same noise spec as the 441. This version was done in October 1997.

Rev 1.2 was done in February 2011. This revision switched the position of R8 (1K) to the output side of R9 (100K) to reduce the voltage divider effect at the output (a very minimal change) and to bring the external 10K gain controlling resistor on to the PCB under the name of R11. Also, the on board high frequency boost resistor (R7) was changed from a 21.5K resistor to a 27.4K resistor to reduce the treble gain......a very small amount

Rev 1.3 was done in October 2011. This revision was to address complaints of picking up radio interference in the preamp. An RF input filter was added to the front end by using R12 (1K) and C7 (220p). This blocks any radio frequency signals from entering the input circuitry of the preamp.

Like Ken has said in many posts and on numerous occasions, there really isn't any big sonic difference between the various versions. I was just trying to address various issues as they came up. I hope this helps get these revision numbers cleared up.

[GeorgeB]

I've always wondered why there is no decoupling cap (100nF ceramic) accross the power pins of the opamp. While the opamps used in this pre are not high bandwidth and thus not very prone to parasitic oscillation with suboptimal supply impedance it would still be common engineering practice to have some decoupling installed, just in case.

[Paul Gagon]

I've always wondered why there is no decoupling cap (100nF ceramic) accross the power pins of the opamp. While the opamps used in this pre are not high bandwidth and thus not very prone to parasitic oscillation with suboptimal supply impedance it would still be common engineering practice to have some decoupling installed, just in case.

Good point George. I always used power supply decoupling on my own circuits when doing any type of preamp circuitry, including the pedals I did for BBE. Leo didn't use the decoupling caps on the preamp and, with the IC chips used, I never saw any issues with parasitic oscillation so I left things alone. The only circuitry I added to Leo's design was an RF filter on the input. The reason for that was because of complaints coming into the sales guys about occasionally picking up radio stations.

However, having said that, you're right. There is really no reason to not put a decoupling cap on the PCB. Of course I wouldn't be able to do that now as I don't work there any more.

[j&lbass]

Hi Paul (and everyone),

This is very helpful, as are all your posts. I would be lost without this site!

I just wanted to add a word of warning for those who are considering replacing their preamp with the one that G&L is selling in their online store. While they promise a USA version and have a photo of same, I bought one several months ago and it clearly isn't the USA version pictured on their site or the updated one that Paul has pictured on this thread.

Here's what they promise (same as Paul's):

img_9991f_1024x1024.jpg (92.13 KiB) Viewed 15116 times

And here's what they sent me (the one that came with the bass is on top):

Preamp.jpg (54.16 KiB) Viewed 15116 times

After getting the wrong part and spending many hours trying to untangle the rat's nest and wrong-colored wiring in my L2000 Trib, I'm losing faith in a pretty excellently designed product. Sigh.

This is just a reminder to double check that you are getting what you pay for.

Thanks,

John

[Ken Baker]

Welcome to BassesByLeo!

Hi Paul (and everyone),

This is very helpful, as are all your posts. I would be lost without this site!

Thank you for the kind words. Much appreciated.

I just wanted to add a word of warning for those who are considering replacing their preamp with the one that G&L is selling in their online store. While they promise a USA version and have a photo of same, I bought one several months ago and it clearly isn't the USA version pictured on their site or the updated one that Paul has pictured on this thread.

I will pass your concerns on to G&L.

Ken...