Voltage Dividers vs Bias

[Littlewyan]

Yesterday when tinkering with my latest creation I learnt just how important the bias of a valve is compared to the voltage divider before or after. I built a 2 channel amp (SLO style, with my own values) and have been battling to get the clean/crunch channel right. The clean sound lacks clarity and crunch mode just never sounded anything other then harsh.
The 2nd stage bias was set with 100k anode and 5.6k cathode, I paralleled another resistor to make the cathode 2.2k and I couldn’t believe how much this changed the sound! I need to do further tests to determine if it’s the right sound, but the clean actually became clean! If I played the start of Civil War by GnR it would always sound like a jumbled mess. With the 2.2k you could actually make out the notes! I should add that I did adjust the volume when doing this to account for the gain difference.

I want others to chime in with their thoughts and findings on this subject. It’s fascinating!

[Stevem]

If your going to get a good grasp on what's taking place waveform wise and infere that into place with what your hearing then you need to use or get a scope!

If you have a given tube biased such that the top or bottom of the waveform is clipping far faster then the other side of the wave then that in and of itself makes for crappy harsh distortion.

This will change a fair amount dependent on the the trans conductance ( how good ) of the tube your working with and can make or brake your effort to get and keep the sound / tone/ play feel your loloking to have!

[Littlewyan]

I thought asymmetrical distortion was supposed to sound nice? And asymmetrical distortion sounded raspy?

[Roe]

was the cathode resistor bypassed by a cap or not? (I guess not)

[Littlewyan]

No, non bypassed. I basically set it up as a JCM800, but with a lower cathode resistor for more gain and then a switch to add in another 470k before the gain pot and a 39k to ground like the SLO100 for the clean.

[martin manning]

Have a read of this if you haven't already: http://www.valvewizard.co.uk/gainstage.html

[Littlewyan]

I've got Merlin's book, great read, but unfortunately I don't feel it really explains why this makes a difference in tone.

I actually tried using my oscilloscope and spectrum analyzer tonight to find out why, but didn't come up with any clear results. Other than very minor (barely noticable) changes in harmonic levels and obvious changes in signal levels, there didn't appear to be any differences. I'm going to try again tomorrow night when I'll have more time.

I do wonder if it's because of the lack of gain. With a 5.6K cathode resistor and only being driven with a tiny 50mV signal, the stage isn't really doing much. Compared to if I used a 2.2K Cathode Resistor and adjusted the voltage divider for the additional gain. I'll update this when I've done further tests.

[roberto]

I suggest you to substitute the cathode resistor with a multiturn 10k potentiometer used as rheostat in series with a 470 Ohm resistor to ground.
This way you'll be able to find the sweet spots for the sound you like. Voltage supply is another important factor as well.

[martin manning]

Your 5k6 and 2k2 Rk produce different operating points and will therefore have different distortion content.
From Blencowe's article: A(unbypassed) = −μ Ra /( Ra+ra+Rk(μ+1))
From that you can see that the 2k2 Rk actually has more gain than the 5k6, almost 6dB more.

[Littlewyan]

Ah roberto, I was wondering if you would chime in. I've been reading a lot of your posts over at slocloneforums and a lot of your posts here with regards to preamp stages. Very informative! Shame some of the graphics are missing, but I still learnt a fair bit. I also read Malcolm Irving's posts and bits on his website with regards to Overdrive. After much reading I did further experimentation with my amplifier on both the Overdrive channel and Clean channel with a scope. For now I'll just attach screenshots of the Clean channel so we can concentrate on one thing at a time.

I input a clean sine wave into the amplifier and ensured that the signal stayed clean on the anode of the stage I was measuring when changing the cathode resistance (I used the 820Ohm first to make sure it was clean, nothing was changed when testing apart from the cathode resistance). I then used the spectrum analyzer to compare a 5.6K to 2.5K and 820Ohm (Resistors in parallel). Remember, I found 5.6K to sound dull, lifeless, servere lack of clarity, when playing a picking pattern it was difficult to hear the individual notes. When I put another resistor in parallel to make it 2.5K, the amp got a lot clearer and was actually a joy to play clean. 820Ohm appeared to be too much, didn't sound very nice anymore, but still clearer than 5.6K.

As you can see in the attached screenshots, the 2.5K Cathode has significantly more Third Harmonic present and a little bit more of the Fourth Harmonic. I then tried making the Cathode 820Ohm and found the Fourth Harmonic increased even further. I didn't realise it would make such a difference given it's got hardly any signal to amplify!

When I get a chance I'll put together screenshots I took of a stage in my overdrive channel before and after adding a large grid stopper.

[Littlewyan]

Sorry Martin, didn't mean to post at the same time! 2.2K does have more gain, but even when compensating with the volume control or my guitar volume control for the extra gain, it still had a much better tone. I wasn't happy with that though, I wanted to find out why! Which is why I got the scope out

I should add that I repeated these tests multiple times to ensure there were no flukes and the results were consistent.

[pompeiisneaks]

From my own side of things I found the same thing.

There was a small epiphone amp a buddy had, probably their version of a champ, very small, low output, but the thing's 'drive' channel sounded pretty anemic. I looked at the schematic and noted the drive stage was using a 100k/2.2k setup, so I swapped the 2.2k to a 1k and man it sounded so much better. Gave it a great gain tone etc.

Sometimes simple tweaks like that go a long way.

~Phil

[martin manning]

Sorry Martin, didn't mean to post at the same time! 2.2K does have more gain, but even when compensating with the volume control or my guitar volume control for the extra gain, it still had a much better tone. I wasn't happy with that though, I wanted to find out why! Which is why I got the scope out :D

I should add that I repeated these tests multiple times to ensure there were no flukes and the results were consistent.

No worries. Looks like your fundamental with the 5k6 is down about 5dB re the 2k2, as predicted. Any reason you are looking at 100 Hz?

[roberto]

On a similar amp I have 100k 4k7 on that stage and 100k 820R on the following, then the CF around 340 V.
That channel is set to be a pushed clean that becomes a crunch when gain passes 7.
It is very dynamic and being mixed with lead (à la Soldano SLO100) gives more nuances to it.

One of my preferred second stages for lead has 5k6 on its cathode (with a lower anode resistor) and it is very very defined.
Another one has 13.3k on the cathode with higher anode resistor and lower supply.

This is just to say that at the end is a symphony of all stages including the speakers, considering how they change their harmonic content while the input signal changes in frequency and amplitude. Then, being a second stage of the amp, you should consider also wider swings on its input.

[Littlewyan]

I've always toyed with the idea of buying one of the those little Epiphone Amps and tinkering. They are basically a weirdly modified Champ. I'm sure there is another made by another brand that has a tonestack in the circuit.

Martin - No real reason, just picked it as a place to start. 1Khz would have probably been a better starting place for my grid stopper tests though!

Roberto - That's pretty much my current setup. Except the 820Ohm Cathode is bypassed by a 0.68uF Cap and the Voltage divider between the 2nd and 3rd Stage is typical JCM800. What divider have you used? I will say that my clean channel did sound good when I split my humbuckers into single coils, but going to humbuckers (wider swing on input) is where it fell apart.

I've made a list of mods I need to try on my amp to tweak the overdrive tone, which I will get round to.