Overdrive - second stage #124

[Stephen1966]

This is an echo of the 124, my Skyliner is different in lots of ways but I thought I would tell you about what I found by trying different values of resistors in the second stage of the overdrive.

I start by saying I am now back at the stock values for the second stage: 180k on the grid stopper and 150k on the output (from the plate). I find this OD is tighter. Not as much gain, a little less bloom perhaps, but it readily starts to feed back under the right conditions.

I tried a 160k grid stopper and a 130k off the plate, the sound was looser, less focused, a little overblown. Loads of bloom, and gain - well, you would be having a laugh if you said this amp lacked gain.

The original option feels brighter, a bit more mid-scooped, and gain isn't an issue.

If anyone else has tried different grid stopper and output resistors, I would be interested to hear your thoughts...

Stephen

[fred.violleau]

Interesting!
Did you record sound samples ?

I would like to document the different build (clean and OD side ), low and high plate, skyliner vs classic etc... so I am assembling a master breadboard where I will be able to switch components to sample sound in the same environment with constants.

It feels like lead dress, OT transformer and component choices bring in many variables and variations on the end result.

We will then be able to choose towards which build our ear is drawn to.

Fred.

[bepone]

depending of the OT (if the OT has low inductance, or bad bass behaviour, all intervertions on lower bandwith in OD are useless),
150k and 100k resistors (with belonging coupling caps 5/10nF and OD pots) can be higher or lower..this will change the character (lower values increasing the harmonics) of the OD and bass in OD..also cathode resistor can be increased to have more sustain and overdrive (colder bias- always good way for better overdrive).

sometimes im thinking for the new amp with new iron to put trimmers there and find optimum value fast what works the best

[Stephen1966]

Interesting!
Did you record sound samples ?

I would like to document the different build (clean and OD side ), low and high plate, skyliner vs classic etc... so I am assembling a master breadboard where I will be able to switch components to sample sound in the same environment with constants.

It feels like lead dress, OT transformer and component choices bring in many variables and variations on the end result.

We will then be able to choose towards which build our ear is drawn to.

Fred.

What a good idea (breadboarding the circuit)! Alas, I tried recording it with my phone but it was so horribly clipped it was useless. It's a simple enough mod if you have the regular preamp board and layout though, and it would be much better if you tried it yourself with your amp voicing and dialed in the way you like to hear it. Lead dress is critical and it's important to avoid putting out of phase lines side by side in my opinion. The photos tell a very interesting and complex story. Component choices, that's where things start to get into a level of subtlety. One resistor is pretty much the same as another, but if you ask me what plate resistors I would use, Dales or TE Connectivity (I tried both), I would say the Dales without a doubt. Why? Is harder to explain Changing the values though, the differences become significantly audible. The topology of the ODS is relatively the same between the versions, so planning the next build, I will be using the Skyliner topology as a start point and building a high plate from it. It's a handful of different components and values but I am expecting it will be quite a different beast.

150k and 100k resistors (with belonging coupling caps 5/10nF and OD pots) can be higher or lower..this will change the character (lower values increasing the harmonics) of the OD and bass in OD

I think I agree with this characterisation of difference in tonal content. "Bloom": harmonics. "Tighter": bass response. I didn't try values as low as 150k or 100k Given that I prefer the sound with the higher value resistors I don't really want to go there. I was getting a rough sounding distortion on single notes with the 130k OD out resistor and back at original values, I have to play it a bit more, I think the 150k has cleaned it up/smoothed it out. What I will look at is reducing the grid stopper back down to 160k and see if that restores some of the bloom (the harmonic content) without distorting the signal out. Thanks, as well for the pointers on the cathode resistor. I might try a 1.8k or a 2.2k there and see how that shapes up. After every single change, I have to live with it a while, to get a good feel for it.

[Guy77]

Another way to experiment with different sounding Overdrives is to change the value of the trimmer pot and the resistor that comes before it at the overdrive entrance after the relay.

Most use a 220k resistor going into a 100k trimmer. Keep the total values the same (about 320k)and mix it up. So you can have a 270k resistor and a 50k trimmer or a 25k trimmer and a 300k resistor.
You will notice the overdrive getting tighter as you lower the value of the trimmer.

Cheers

Guy

[bepone]

I think I agree with this characterisation of difference in tonal content. "Bloom": harmonics. "Tighter": bass response. I didn't try values as low as 150k or 100k Given that I prefer the sound with the higher value resistors I don't really want to go there. I was getting a rough sounding distortion on single notes with the 130k OD out resistor and back at original values, I have to play it a bit more, I think the 150k has cleaned it up/smoothed it out. What I will look at is reducing the grid stopper back down to 160k and see if that restores some of the bloom (the harmonic content) without distorting the signal out. Thanks, as well for the pointers on the cathode resistor. I might try a 1.8k or a 2.2k there and see how that shapes up. After every single change, I have to live with it a while, to get a good feel for it.

180k grid stoper has several functions , it is creating hi freq. roll off with input grid capacitance + during heavy overdrive it is creating clipping with g1 current which starts to flow in g1. so increasing it will lead to overall smoother response but also will create more distortion in higher volumen/drive.
increasing that stoper is also ok.

150k and 100k also have several functions, lowering them will lead to tighter bass (110-130k) but also in the same time more distortion/harmonics due to increasing the total active load to preceeding anode which is driving that.

1.5k->2.2/2.7k in the V2b cathode (without touching anode resistor) will change bias for this stage and significantly increase sustain in the stage, overdrive in that stage (more to clipping and finally fuzzy sound) which can increase quality in OD!

[Stephen1966]

Thanks Guys, I'm working on it now. For the moment, I have a 220k going into a 100k trimmer. I would say, that is about right here because that bass has a lot of EQ possibility on the front end and with various pickup setups. The trimmer also has a say in the bass response... I haven't played with the trimmer yet. I want to keep it where it is, which is about 40k to ground and keep that a constant to keep things simple. It's an interesting idea though.

180k grid stoper has several functions , it is creating hi freq. roll off with input grid capacitance + during heavy overdrive it is creating clipping with g1 current which starts to flow in g1. so increasing it will lead to overall smoother response but also will create more distortion in higher volumen/drive.
increasing that stoper is also ok.

So by dropping this resistor back down to 160k that should lower the cutoff frequency and reduce the distortion.

1.5k->2.2/2.7k in the V2b cathode (without touching anode resistor) will change bias for this stage and significantly increase sustain in the stage, overdrive in that stage (more to clipping and finally fuzzy sound) which can increase quality in OD!

This is worth investigating, I believe, I think I can try several values side by side soldering a couple of pins in and jumpering the resistors with clips. More sustain, sounds good and a cooler bias might just do the trick.

[Stephen1966]

I tried a 1.5k alongside a 1.8k, 2.2k and 2.7k cathode resistor on the second stage of the OD. The 2.2k does the trick

I really wanted to get rid of the distortion on the treble side and while the 1.8k made a noticeable difference, cleaning the sound up, and the 2.7k had a more pronounced smoothing effect, it was the 2.2k that represented the best of the bunch. The 1.5k was of course, my reference.

There are three reasons why I think the 2.2k is the best. First, it cleared away that distortion, effectively the triode needed a cooler bias. Second, the sustain is probably a little better though if I am being honest, there wasn't much difference between the 1.5k and the 2.7k. I mean, to the point where I really couldn't tell. But the third reason is because it still has the rich, upper partial, velvety tone. There was a distinct metallic ring with the 2.7k which I attribute to the onset of fuzz and that really wasn't to my liking though it's one of those things, that if you didn't know any better, you would probably find it quite acceptable. If that is the kind of OD you like, then go for it. But I like the rich mids and that kind of smooth OD. These changes I feel hollowed out the bass frequencies a little as well, which isn't unpleasant because when I restored the 160k grid stopper, I got a nice augmented projection of bass frequencies I could hear, a kind of swirly, more harmonically rich content. It dialled back the OD distortion (for which, read "grit") a tad as well. Which again, is not an unpleasant effect because the Level and the Ratio of the OD were but at midnight through all these attempts and I might dial up the Level and dial down the Ratio for a grittier OD if I need it. I can also start to play with the trimmer on the OD input now as well.

I am guessing that if such a thing as a 345k trim pot was readily available, we could have achieved more or less the same results with the trimmer but when you run the signal through a 220k resistor followed by a 100k trimmer we need to think about compensating for the effect on the lower frequencies it has with the grid stopper for the second stage and it cathode bias. Fascinating workaround. And all quite easily implemented. Thank you Bepone for the hint about the cathode bias. That was the key, I was looking for.

Edit: I took a photo of the testing set up:

IMG_20220719_192230.jpg

Out of curiosity, I ran some figures on the difference between a 160k grid stopper and 180k under different load conditions: with a µ of 25, 50, 75 and 100 (the last being the theoretical upper limit to the gain of an ideal tube). These are just numbers. But I was interested to see what difference the Miller capacitance has at different levels of gain when forming a low-pass filter with the grid stopper (Rg). I'm assuming the interelectrode capacitances are Cgk = 1.6pF and Cga = 1.6pF and the formula I'm using is: (the input capacitance) Cin = 1.6 + 1.6 x (µ + 1).

This gives the following:

µ = 25; Cin = 43.2pF
µ = 50; Cin = 83.2pF
µ = 75; Cin = 123.2pF
µ = 100; Cin = 161.6pF

I input these figures into a low-pass filter calculator.

For a 180k Rg, the frequency cutoff (Fc) - rounded up and down:

µ = 25; Fc = 20.4 kHz
µ = 50; Fc = 10.6 kHz
µ = 75; Fc = 7.2 kHz
µ = 100; Fc = 5.5 kHz

For a 160k Rg, Fc:

µ = 25; Fc = 23 kHz
µ = 50; Fc = 12 kHz
µ = 75; Fc = 8 kHz
µ = 100; Fc = 6 kHz

So, if I have all my ducks in a row here, this says the lower value grid stopper has effect of elevating the cutoff frequencies, or allowing higher harmonics to enter the signal. All of these frequencies are well above what the guitar can produce so the interesting next question is whether it has any effect on the presence of even-order harmonics; those we consider pleasant, that have consonance. Most people's ability to hear higher frequencies top out around 20 kHz so for practical purposes, most of these levels of gain and the harmonics they produce on top of the fundamentals in the signal, are within an audible range. I'm happier with the 160k grid stopper than I was with the 180k but you would have to listen to it in a purely acoustic space, not one that can easily be recorded, to perhaps understand why.

[bepone]

There are three reasons why I think the 2.2k is the best.

Thank you Bepone for the hint about the cathode bias. That was the key, I was looking for.

you're welcome! i tested also hi plate amps in the same way, 220k anode / 3k3 cathode/ where i finished with 4k7 in V2b cathode!

[Stephen1966]

There are three reasons why I think the 2.2k is the best.

Thank you Bepone for the hint about the cathode bias. That was the key, I was looking for.

you're welcome! i tested also hi plate amps in the same way, 220k anode / 3k3 cathode/ where i finished with 4k7 in V2b cathode!

Cheers! I have to imagine that produces a very focused, tight driven sound.