London Power ad

Search the Forum

(Advanced Search)

Preamp mods
Hi Guys

There are as many potential modifications that can be made to a tube guitar preamp as there are players. Most mods are pretty simple, like adding a cap or changing a cap value, where others are more extensive, like adding complete new channels of circuitry.

Whatever you are interested in doing, we can discuss it here.

Have fun
Greetings Kevin,

I am currently finishing up my CAD file for my chassis drilling layout, wondering if you could suggest some mods for the LP and Z preamp (3 channels 50w amp, with BFX and Power Scaling). What I have in mind (if its possible to do):

- Add a gain/drive control on the clean channel, don't really see a good spot on the schematic for a pot beside right at the input of V1A or the other would be like the one in my friend's Weber Halle, a linear 5k pot on V1A's cathode, wondering how you feel about such a pots placement in the circuit (I only want to add one pot)

- Add some switches to all three channels (bright cap bypass, gain boost/tame, low end boost, ect). I was thinking about having three switches per channel, if its possible.

I just want to have as much flexibility as possible with this amp, kinda want it to be my ultimate studio weapon.

Cheers and have a good one!
Hi Vostre Roy

There are countless things you can do with LP-PRE as TUTs show, where it is the LPSP - London Power Standard preamp.

The "low hanging fruit" of options:
>> make each Ck switchable
>> make the attenuator high-frequency emphasis switchable
>> add a bright cap to either/each channel and make it switchable
>> make the lead-channel interstage attenuator variable

Ck out of play reduces broadband gain for the first and fourth stages, and reduces high-frequency gain in the second and third stages. Making the interstage attenuator variable if done properly only reduces gain - there is a sweet spot of attack before hitting grid-rectification where the attenuator values are set at the moment.

Adding a Gain control to the clean channel has what goal? Do you want to be able to reduce gain? Increase gain? Get an overdrive tone from the clean channel? There are actually a lot of options, most of which would involve having bypassable or liftable controls, or modifying LP-PRE to be like the Ultra-Low-Noise version of LPSP shown in TUT. This would let you move the clean EQ and have a Gain between the clean channel stages and a Level after the EQ. Only mild overdrive is possible unless you add a bounding network or Power Scale the second clean stage.

Other mods are to add a bounding network to the Lead channel, or change the plate Rs or cathode Rs - switching either of these live will cause a POP!

You might want to build the preamp per the note first and try it out. This gives you a base line to work from, and you can make a list of tweaks to do, or features to add "need more of this", "need less of that".
Hey Kevin, thanks for the reply!

Yeah I think its a good idea to build it stock to begin with, but knowing that it is possible to add such features, I'll prepare my drilling pattern so everything is drilled before I begin to actually assemble it. I just don't want to have to come back once the amp is completed to drill extra holes.

I'll make a thread soon about the build, might need your opinion on the stuff placement, right now I've got two layouts but I ain't sure which one would be the best

I have a question about cathode followers.
Kevin says it can be bad for sound, it looks like the spectrum is not so wide anymore.

Does this only play when you feed a tone stack with it?

In the BFX and in say the Sigma a cathode follower is used as a send.
Will that narrow the sound too?

Also, I tried making a Tweed Mixer by adding a triode, with a shared anode of course.
I supposed the gain would be the same but there was a huge increase in volume.
Is that normal when you add a triode that way?

Hi Strelok

I never said the cathode follower narrows the bandwidth - in fact it should widen it.

As a buffer for a send out a CF is fine. It does exactly what a buffer is supposed to do. As a tone-stack driver the CF kills tone. You have to remember that the original tone stacks in Fender amps et al were plate-driven and tweaked to work as desired with that high drive impedance. Reducing the drive impedance changes how the controls work, to the point where the tone stacks function poorly - tone change versus control sweep is greatly reduced so what is the advantage?

The cathodes of the tube sections in a tweed mixer (anode-coupled mixer) must be separate. This allows each input to have its own gain, reasonably independent of the other inputs. You can even mix different tube types.
I agree with Kevin when it comes to driving the EQ with a cathode follower. After reading TUT5 I conducted my own tests on a JCM-800 type pre-amp that was modded for extra gain and 5F6A Bassman circuit. With the plate drive the eq worked much much better. There is a tonal difference but I personally preferred the plate drive in the JCM 800 which sounded smoother and more pleasant. In juxtaposition I've also tried the same thing on 5F6A Bassman type amp and while the same effect of better EQ sweep occurred but I preferred the cathode follower's sound. However, that might have been just because I was so familiar with it's sound. With additional tweaking I probably could have arrived at a set of component values that I liked the plate drive better for.
Hi Folks,

Thanks for all this info!

Makinrose you actually tried it!

Kevin must have experimentied with it too I think.

From your adventures it looks like it is a bit like Baxandall vs Fender Tone Stack, to my ears Baxandall sounds very good on clean sounds but I like Fender better for distorted sounds.

How does our friend over at Orange do what he does?
It looks like a Baxandall stack judging by the one control.

I think the Micro and Dark Terror sound fantastic, never harsh but rich in harmonics and a nice sag to the sound to boot!
How does he create that, wat topology does he use?
I have no schematics.

Kevin, are you going to do a project on Orange?

Kind regards,

Hi Kevin.

Regarding the LP-PRE pcb, are there more relay positions available on the board, in addition to channel switching, for items discussed above?

Like Ck and such.

I want to get the number of switch components correct on order.

Thanks very much.
Jim Gale
Hi Guys

Strelok: Seems I missed your message for a while - oops!

I like Baxandall EQs and use them in my amps. The control interaction is minimised and you can add a middle pot in a couple of different ways.

When there is only a single tone control, it can be implemented several ways. Yes, it could be half of a Bax - just pick if you want to control lows or highs. It will be passive and the net result is the same, skewing the response to be "bright' or "boomy".

You can also strap a control between the outputs of a low-pass filter and a high-pass filter, as Electroharmonix did for their "tilt" control. It works well. I believe Duncan's EQ sim has that.

You can also just wire a tone stack but use fixed values for the bass and mid positions - very common. Or go even simpler as in the earliest amps and in guitars, with just a cap and pot to dump treble to ground.

Galejt: LP-PRE just has the mute connection blocks for channel switching.

Considering the tightness of the component spacing, it would be best to add a second board to mount the switched components on, maybe as an eyelet board, or whatever makes changing the optioned part for a different value.This way allows you to separate the final switch means from the experimental switch means, and from the experimental phase itself.

Have fun
Hello Everybody!

Thanks for the info Kevin!
Maybe I'll try an active filter like in your books.

Yes I saw that first on a Rectifer Preamp, at the second stage a completely hardwired Tone "Control".
Having the trebel as a pot could do wonders.


Hi Guys

Strelok: Earlier on you suggested that you changed the output stage of the London Power Standard Preamp to a tweed mixer and got a huge loudness jump. The reason for that is that there is no longer attenuation at the output of the Clean and lead paths.

At the input of the LPSP 4th stage, two 470k resistors tie the Clean and Lead signals to the grid. The source end of one resistor or the other is grounded, allowing the opposite path to be active. This muting action creates a voltage divider, cutting the enabled signal in half at the grid.

When you replace the 4th stage with a Tweed Mixer, each path feeds directly into a grid (via a grid-stop, hopefully) and there is no specific attenuation. Grounding one grid or the other does not alter the signal level of the enabled path, resulting in a signal twice as high compared to the stock LPSP.

Note that if you use a relay for this selection, it is unlikely that any series resistance need be added ahead of the grid-stop. If you use jfets, then series resistance will be required as the jfet Ron is not zero.

You can use higher Rk values for the two triodes of the mixer. 1k5 is getting you extra loudness; 10k would be a very low gain of 3x through each path; values in between would achieve intermediate gains and loudnesses; likely something around 3k3 would be close to the stock signal levels. As I mentioned earlier, Rk values can be different to achieve whatever signal balance you desire.

Have fun
I only know simple amps. But for a Tweed Fender champ style amp, I was thinking of sacrificing one of the inputs to put a switch, that would, connect to the cathode of the second half of the 12ax7.
Position 1. add a bypass cap on cathode (by attaching the cap to ground)
position 2. attach negative feedback.
Position middle. do nothing

This is not my idea of course. And it might be advantageous to build it in an external box. That way I can see what I like by clipping it in the right places, and then decide which to do. I mean, often times I find what I like on an amp and don't change often.
I also often think that I would like to hook up a rotary switch for plate resistors...or cathode resistors. Maybe mount it on the transformer side of the 5F1 chassis, so no one will be tempted to try and switch while it's running. (hidden) But that's got to be noisy... I'd be afraid that it might arc across it. Maybe one day.
Hi Guys

Drew - All of those ideas can be tried and have been tried by many tinkerers and builders.

First, regarding repurposing the second input jack: The hi/lo jacks on Fenders are fairly redundant now, since every player has his own amp and the low-gain input is also low-z, which is not so great for your guitar tone. All the extra jacks make great spots to add new pots or switches as your mods dictate.

When there is also a separate triode related to each jack, there is a world of change possible by re-ordering the gain stages, ie, place them in series instead of parallel. This is what turns a plexi into an 800. Even for clean playing, excess gain can make the sound more "liquid" and it can feel like the guitar plays itself.

Regarding switching plate / cathode resistances: There will be a DC shift when these paths are interrupted, so you may want to turn a later volume control down before making the switch. Otherwise, you can make the selection "live" with power on. The way to minimise the thump is to always have a high-value R that remains intact as Rk and as Ra, then add Rs in parallel with them. This is "shunt" switching and it is always quieter than series switching.

An alternative is to have many low-value Rs in series, then the switch shunts the sections not needed. Just as above, there is a DC path that remains intact regardless of the net value selected for Rk or Ra.

Note that in general, when you make mods to the amp that effect DC paths the power should be off.

If you make a mod to the bias supply, REMOVE THE POWER TUBES and test that bias voltage appears right at the tube sockets and is of sufficient magnitude to control the tubes. Then turn power off, install tubes and proceed with bias setting.

Have fun
Wow. Awesome knowledge Sir. So essentially put like a 1 meg resistor at the end of the bypass capacitor and use the switch to put the ground where it meets the capacitor to "activate." As for the negative feedback, I take it a similar setup? Then no pops? You're blowing my mind dude.
Hi Guys

Drew, it would be easier with a schematic of either the starting circuit or the result you want to get to.

To address the last post first: Isolating a cathode bypass cap (Ck) requires a resistance in series with it that is 10x the Rk value. Going higher than this in the form of a pot makes most of the sweep of the pot ineffective.

Keep the top of the cap tied to the top of Rk and switch or dial the negative end of the cap to ground. If using a switch, add a resistor across the switch to eliminate thumps - again, 10x Rk min for the new resistor.

Feedback around tube stages: note that each triode GAIN stage is inverting, and there is only one (general) way to apply feedback around a single stage. There is already feedback around each gain stage in the form of internal capacitance of the tube along with parasitic capacitance from wiring. There are in picoFarads, but they roll off very high frequencies. TUT Fig.4-8 shows the frequency response change made by this capacitance as it works against the grid-stop resistor.

If you add a high-value resistor from anode to grid on a triode stage, the new R works along with the internal capacitance to form another high-frequency roll-off. You could use a low-value cap instead and have predictable high-frequency filtering, good for an input stage. The feedback resistor or cap can be tied in after the plate-coupling cap to avoid DC issues. Similarly, the cap can be a shunt to ground, simply loading the triode at high frequencies.

Feedback around two triodes provides reasonably good gain control, better applied in bass amps than for guitar. Generally, guitar players want the harmonics from free-running stages working at medium to high signal levels with resp[ct to the stage's transfer curve for its given voltage environment. The absolute voltages do not have to be especially high, but everything factors in to the sound you want.

Forum Jump:

Come in where it's warm!
A warm welcome to tube amp modding fans and those interested in hi-fi audio! Readers of Kevin O'Connor's The Ultimate Tone (TUT) book series form a part of our population. Kevin O'Connor is the creator of the popular Power Scaling methodology for amplifiers.
Please remember these three principles: respect, sharing, community.
Not familiar with The Ultimate Tone book series? See discussion topics, or click here to visit London Power/Power Press Publishing.

Tube Amp Forum Hosted by London Power
London Power logo