midrange and how to get it

[tubeswell]

Look at a middy sounding lead amp like an SLO100 - The first four stages have got all sorts of mid stuff overdrive (the more stages you have the creamier you can get it)

.02uF couplers
1uF/1k8 cathode combinations to roll-off bass and emphasize mids (and highs)
lots of interstage attenuation to help get rid of raggedness (and highs)
a cold-biased stage at the end to get creamy sustain - with 39k cathode resistor and a treble shunt plate bypass cap
220k plate resistors here and there for ott gain - even on the DC coupled pairs

I copied this into my C30 and the OD channel sounds pretty good now. Can it do mids? very well Does it need to do other stuff? Nup

[daz]

I don't have enough stages to do a cold biased stage. Tried it once like a JCM but i then didn't have much drive left at all after that. I do use a lot of similar stuff tho. .022uf couplers, three .68uf bypass caps (maybe i should try 1uf? tho i have and .68 seemed better, at least back when i tried it) 220k plate at v1a.

One of my problems that may account for a lot of my endless tweaking is that i rarely play gigs anymore, (which is the best way to evaluate) and the only 2 places i can play and tweak it are acoustical nightmares. For example, here at home it's all hard surfaces in a small room. Tile, hard walls, glass. So with that in a small room there are so many fast reflections i can't get any articulation, so i never know how close i've come to accomplishing that. And clear articulation is one of my goals. It's tough.

[Enzo]

daz, before you go too far comparing the brands of caps, you need to MEASURE the capacitance of each. E-caps have much looser tolerance than the other parts we use. You could put two different branded 20uf caps into something, and while the labels both say 20uf, one could really be about 16uf and the other 24uf. That is a considerable difference right there.

Formulations differ, so if you measure the same capacitance, then differing ESR and other factors enter into it, and that starts to show real brand differences.

[daz]

I just checked the 16UF atom i replaced the F&T with and the F&T. There is a 5UF difference between them....atom=20uf, F&T=25UF. Funny....i have two atoms, one 20 and one 16, and the 16 measures 20. According to chuck that shouldn't be much difference tho. The atom LABELED 20uf seems to have found it's way under something where i can't find itto measure it. In any case, the tonal change was not subtle at all, so i have to think it's not the cap value that caused it. I have heard of ESR but i'm not sure what it is or how it affects things. But regardless of what causes it, it's now apparent different filter caps DO make as difference, and it CAN be more then subtle.

[Chuck H]

I have heard of ESR but i'm not sure what it is or how it affects things. But regardless of what causes it, it's now apparent different filter caps DO make as difference, and it CAN be more then subtle.

Equivalent series resistance.

It means there is a measurable resistance across the cap, so...

Since these caps decouple AC to ground any resistance causes a small defeat in this function. This means that there are many tiny FB loops that do effect the AC signal path lurking in the power supply rail. Some of these loops are NFB and some PFB. Exactly how this phenomenon effects the sound of the amp depends on how much resistance there is at a given frequency and what decoupled circuits share filters. For example, V1a and A1b typically share a filter. If V1a's plate is out of phase with V1b's plate, if they are cascade stages for example, the effect is NFB. If they are parallel stages, one input amp and the other an effects loop or reverb driver for example, the effect is PFB. Since the ESR of the filter cap is much lower than the B+ rail series resistors I don't think there would be any significant effect from adjacent filters. But any shared filter does have this effect. ESR at different frequencies varies with construction (brand) and overall value (uF) so different caps do sound different for better or worse. The excact effect will differ depending on your filter ground scheme. This is one reason why it's important to follow a layout when cloning a specific amp. It also means that there is no right, wrong or best filter type since a poor filter (typically of higher ESR) could have a desireable effect in the right circumstance. I don't know any builders who have taken this to science and learned any defintive "good/bad" rules around this phenomenon either. Probably because many amps supply rails have different stages sharing filters. It's just too many variables.

Observations

Chuck

[daz]

Thanks Chuck. Learn something new every day. (and i change my amp into something new nearly every day LOL!)

[daz]

Tubeswell....thanks for mentioning that...i copied the 1.8k/1uf cathodes on V1A and B and the tone is a definite improvement. I played the amp with a cranked CD to see how it fits in the mix and A/B'd the R/C combos i had and the soldono values and w/o a doubt i like the 1.8k/1uf values on both sides better. the only thing is i'm wanting a tad bit of very high end sizzle on the end of the distortion to bring out the harmonics and cut in the tone. Any ideas? Not that it's bad, it's better. Just that it would be perfect with that, and it's so so close right now. Funny...i always looked for ways to remove that fizz, but with this combo a bit of that would be perfect.

[SysCrusher]

You need to think about how treble peakers work like what Marshall did with their typical 470k/470pf in parallel with 470k to ground. If you add some sort pf cap to ground after it, your dumping what ever high frequencies determined by that extra pf cap to ground and the preceding 470k that is in parallel with the original 470pf cap. Hopefully I didn't confuse you.

Basically you can peak the mid frequencies, like setting your graphic eq with the shape of a frown centered at 1k, with just 2caps and 2 resistors.

Hint: Marshall set their frequency response between stages with about -6db difference between 10Hz and 720Hz and another -6db between 720Hz and 1000kHz. Most will cut those peakers out but you loose that mid crunch when you do that. Just add pf cap from 250pf to 1000pf to ground to tame the treble peak some.

[daz]

No, you didn't confuse me. And I've done all that with no great success. Right now i'm experimenting with just using 1uf bypass caps at 3 cathodes instead of the .68's, and it's sounding much better.

[SysCrusher]

No, you didn't confuse me. And I've done all that with no great success. Right now i'm experimenting with just using 1uf bypass caps at 3 cathodes instead of the .68's, and it's sounding much better.

.68 is pretty much peaking the treble with out the mids. 1uf will get you more of the mids. Take into consideration what the cathode resistor value is also if you want to pin point frequencies.

To much to think about. Even the plate resistors will effect the frequency range a bit. The math helps to get you close but I swore I would never do that kind of math after college.

[daz]

I think i hit the jackpot on this midrange thing. I tried the 1uf bypasses on V1 and 2 after tubeswell mentioned the SLO using those values. But while i liked it, it lacked a lot of what i had before. So i went back to the way it was which was easily the best it has ever been with .68uf's on V1a dn b. But i also has a .68uf bypassing the 820R at the 3rd stage, and i thought that this would be the best place for the 1uf the same way the EQ is most effective after the distortion. So I tried replacing that .68 with a 1uf and Bingo ! Even tho it's only bypassing a 820R it made all the difference. the tone still had everything i loved about the .68 caps on V1 but the fizzy top smoothed out a bit and the mids took on that round fluid tone i was looking for w/o sounding the least bit nasally. This is just what i've been after forever. time will tell tho, and you all know me. But really, i have truly for the first time nailed an important aspect of the tone i have been after since i built this thing thanks to that one little cap. this is a good day !

[Chuck H]

Even tho it's only bypassing a 820R it made all the difference.

Just as important as the amount of effect is the frequency being effected. Maybe more so in this case. The 820r resistor is a lower impedance circuit. That means that the standard .68 bypass cap you had there was only boosting higher frequencies on that stage than the same .68 value used on a cathode with a higher value resistor. The impedance of the circuit effects the frequency knee of the capacitor. So by upping the value of that cap your getting closer to what was probably the original intended effect. Try other values there too, like 1.5uf and 2.2uf

Chuck

[daz]

i'd try higher values Chuck, but the thing is that i'm finding a 1uf has a certain cutoff thats exactly what i was looking for. And the further back after the distortion it is the better. So if i tried a 2.2uf or whatever i'd have no other place to implement the 1uf, and that one needs to be somewhere to get what i'm after. And thats the only place to put it after V1 where it tends to work but with side effects.What i was thinking of trying is a filter like those 470k/470uf marshall ones but with a 1uf instead of the 470p. problem is, i don't have a 1uf non electro, and those are hard to find.

[imaradiostar]

I had a few thoughts- dunno if these will be helpful or not.

I think the Bogner Shiva can be adjusted for a lovely midrange lead sound- it sounds like some of the mods you've made make your amp more similar to the Shiva. Perhaps there are other secrets to be had there. The schematic can be found on the web without too much digging.

My favorite "thick midrange" sounds come from a pretty specific signal chain. Either a tube driver pedal or tube works rack mount "starved plate" distortion into a cranked up plexi style marshall with enough power tube distortion to be noticably dirty even with the distortion pedal off. I use a THD Hotplate or a home-made attenuator so I can crank it up without blowing everyone away. The channels on the marshall are bridged (y cable into top inputs), bright volume around 11 o clock, normal between 9 and 10, other knobs set to taste. The real trick is to set the treble control on the tube driver or real tube rack unit to 0. Seriously. When the control gets to zero it drastically changes how the controls interact, making for a really interesting tone. I usually set the bass differently for different amps and speakers and the gain and volume to match whatever guitar I'm using. Tube selection seems to be important on the starved plate pedals too...

Anywho, I'm a big Eric Johnson fan and the above is not too far off from what he's done for a number of years- obviously that's why I landed at that combo. For many many years of gigs I've never failed to get a complement on the sound. I'm not a great player really so something must be working there! I'm sure there are other ways but I've used a rack mounted real tube for the last 5 years as my main distortion, always with the treble on 0! Before that I used a four knob tube driver but I got rid of it when I saw how much people were willing to pay for them. The rack mounted unit I have has an added "cathode resistor" to adjust the bias on the tube so I find it's easier to get a good consistent sound with it. Supposedly Chandler is now selling a four knob tube driver with a similar feature.

I don't own any EV speakers but I've been using 70's greenbacks for about 12 years. They're pretty sloppy but man, the midrange when they're cranked up is nice. I just wish they had tighter bass!

jamie

[daz]

Actually i long since got this sorted out perfectly. First, i came to realize while i love that tone, i don't want it all the time. Only as a sort of effect or lead tone for certain things. But i found a better way to get it thanks to someone at the weber forum. I have one of those 470K/470pf filters like JCM 800's use right before the gain pot. It gives me the best all around tone. But if i bypass that and put another one going into the 3rd stage but with a .0047uf instead of a 470 pf, i get exactly that mid tone i was after. I don't know why i never thought of doing that before because i've used them with the 470pf many times. But i never thought to alter the amp's resonant frequency in the dribe stages like that. So i uninstalled a DPDT switch which in one position bypasses the 470pf one leaving the .0047pf one in place and i get that mid tone. Flip it the other way and that one gets bypassed and the 470pf one is in play. I use that almost all the time. Bu i like having it for when i crave that mid tone. But like i said, i came to realize i don't want that tone often like i thought i would. I remember loving that tone years ago, but i was younger then and my taste in tone has changed a lot.