Rootz
Here another member talks about this compression as well. His name is Max
Max
The Dumbleland Specials I had a chance to play all had a very different character as a musical instrument. One was like the best 59' Bassman you can think of on steroids (would perhaps have been SRVs "dream amp"?), one very similar in sound to a SSS without reverb, one (obviously tuned for bass guitar) was different again, and the "Silver-Aligator" "Pentode-Dumbleland" (the one that came back to US from Japan and was sold at Overland Express) was the "sustain master" (After hitting a chord you always looked around and asked yourself "where is the compressor"?). Ok, I will come back later to these differences in sound.
Brandon
There is no compressor these amps have an unexplainable phenomenon! there is nothing strange in them at all. I think it's Dr D's mastery of lead dress and component selection, otherwise it's all regular old circuits.
I have no idea why it sounds the way it does, there was nothing out of the ordinary in the best sounding one I heard, and Electron Avalanche posted a schematic of the "Pentode Amp" preamp here: https://tubeamparchive.com/viewtopic.ph ... sc&start=0
This "Pentode amp he and Brandon talks about DC driven w/ 2 7025's SVT style and does not have the Tweed mixer like #002 C.F.AFAIK..Here it is https://www.youtube.com/watch?v=ftRMqJaAFyI
Here is the best example of it I can find.You can mostly hear it in the top end or when he pops the low E string and the slower attack time..SRV's has a right hand of lead and still the amp does not break just compresses (SRV's Silver SSS has AC Coupled SVT style driver 12AU/12BH7 not sure of the Reverb) .It's hard to explain you just have to play one
Let us know if you find out where this compression effect comes from!! https://www.youtube.com/watch?v=VlDl2JQbMaY
BTW I agree 12AX7' suck for P.I drivers due to the low current handling..
Tony
Last edited by talbany on Sun Apr 11, 2021 12:29 pm, edited 1 time in total.
" The psychics on my bench is the same as Dumble'"
Stephen, those send and return points in Larry's schematic are the effects loop. The reverb recovery stage is in the REV. box.
How about the solution in the schematic I attached? When you switch of the reverb, the output of the OD relay goes directly to the master volume. The only thing this idea doesn't account for is the difference between lead dress between a stock ODS and with an added reverb. Note there are two traces: one with reverb, one without (bypassed). Can you see which is which by the trace alone?
talbany wrote: ↑Sun Apr 11, 2021 11:58 am
Rootz
Here another member talks about this compression as well. His name is Max
Max
The Dumbleland Specials I had a chance to play all had a very different character as a musical instrument. One was like the best 59' Bassman you can think of on steroids (would perhaps have been SRVs "dream amp"?), one very similar in sound to a SSS without reverb, one (obviously tuned for bass guitar) was different again, and the "Silver-Aligator" "Pentode-Dumbleland" (the one that came back to US from Japan and was sold at Overland Express) was the "sustain master" (After hitting a chord you always looked around and asked yourself "where is the compressor"?). Ok, I will come back later to these differences in sound.
Brandon
There is no compressor these amps have an unexplainable phenomenon! there is nothing strange in them at all. I think it's Dr D's mastery of lead dress and component selection, otherwise it's all regular old circuits.
I have no idea why it sounds the way it does, there was nothing out of the ordinary in the best sounding one I heard, and Electron Avalanche posted a schematic of the "Pentode Amp" preamp here: https://tubeamparchive.com/viewtopic.ph ... sc&start=0
This "Pentode amp he and Brandon talks about DC driven w/ 2 7025's SSS style and does not have the Tweed mixer like #002 C.F.AFAIK https://www.youtube.com/watch?v=ftRMqJaAFyI
Here is the best example of it I can find.You can mostly hear it in the top end and the slower attack..SRV's has a right hand of lead and still the amp does not break just compresses (SRV's Silver SSS has AC Coupled SVT style driver 12AU/12BH7 not sure of the Reverb) .It's hard to explain you just have to play one
Let us know if you find out where this compression effect comes from!! https://www.youtube.com/watch?v=VlDl2JQbMaY
BTW I agree 12AX7' suck for P.I drivers due to the low current handling..
Tony
Hi Tony, I'm not disagreeing with the compression effect at all. I hear it, Max does, you do and a lot of other too probably. It just is there. I just can't explain it from a 'current stealing' point of view, because AFAIK AC coupled cf's don't steal current. I take it that SRV played his amps very loud and compression can come from that alone, but in the clips here the amps are turned down and you can still hear it. I hope I'm not jumping to conclusions, but I think I (we) are missing something here.
talbany wrote: ↑Sun Apr 11, 2021 4:36 amClass A2/AB2/or B2 requires a very low impedance, transformer-coupled or DC-coupled driver stage. The standard AC-coupled phase inverter or single-ended driver stages used in nearly all guitar amplifiers will not allow grid current flow, so they are class A1/AB1/B1 amplifiers.
The advantage of class A2, AB2, or B2 is the complete lack of "blocking distortion", or transient intermodulation distortion. The disadvantage is the extra complexity of the output stage required to source current to drive the output tube grids into the positive region.The main advantage to the cathode follower directly into power tube grids is losing the effects of grid-blocking from the coupling caps recharging and causing momentary distortion when power tube grid current occurs
I think elimination of blocking distortion might be the main reason for the CF drivers, rather than the potential output power increase of operating in AB2, although the 7025 drivers can manage to supply a few mA of grid current.
As Rootz points out, the CF's in 002's mixer stages are DC coupled to the voltage amplifier stages feeding them, which is the situation described by Blencowe. The DC coupled CF power stage driver in SSS 002 is the opposite, DC coupled to the output tube grids and AC coupled to the PI. The driver stage bias point is at around -4V, so nowhere near any continuous grid current flow.
rootz wrote: ↑Sun Apr 11, 2021 12:25 pm
Stephen, those send and return points in Larry's schematic are the effects loop. The reverb recovery stage is in the REV. box.
How about the solution in the schematic I attached? When you switch of the reverb, the output of the OD relay goes directly to the master volume. The only thing this idea doesn't account for is the difference between lead dress between a stock ODS and with an added reverb. Note there are two traces: one with reverb, one without (bypassed). Can you see which is which by the trace alone?
If we are looking at the same schematic (Larry's) I don't see a separate trace - one with and one without reverb. In your simulation screenshot however, I do see a separate trace, from U101 to the PI in the bottom schematic without reverb. Whereas in the top simulation (and the middle) the master appears to be after the mixers coupling cap... U64 top, U65 middle.
Anyway, I have an alternative switching arrangement here:
ODS RVB 2.9.pdf
This bypasses the reverb and goes directly from the OD relay > RVB relay > Master > FX loop > PI and switching on the reverb it goes OD relay > RVB relay > RVB > Master > FX loop > PI. The only difference here is that the static line between the input of the RVB circuit and the master after the output is broken by the RVB relay... perhaps that was the original mistake that needed to be fixed?!
You do not have the required permissions to view the files attached to this post.
I think that is the point of the design of the SSS: maximum clean headroom, back when stage volume was everything. So, without farting out.
And that was exactly what I mean Martin. There is AC coupling to the PI (the side that counts when talking about 'stealing current'). I don't think there is much compression at all if any in the cf drivers. Why? 1) high input impedance (which actually means the PI will compress less; lighter load), 2) probably (will have to do the math to be sure) nowhere near grid conduction, 3) 100% negative feedback, so highly linear by nature.
But then again, what causes the compression? Might it be the stages with local negative feedback. I could see those feedback networks compress when the input approaches grid conduction. I the preamp, with the master just in front of the PI, there easily are conditions where the AC voltages at the grid would be high enough to cause clipping and/or compression. Again, that also is in case of high enough voltages (turning up the gain, playing hard, PAB) and I think I hear the compression all the time. Martin, Tony, is there much compression when playing softly? When does it happen?
Martin, do you have another explanation for the compression?
rootz wrote: ↑Sun Apr 11, 2021 12:25 pm
Stephen, those send and return points in Larry's schematic are the effects loop. The reverb recovery stage is in the REV. box.
How about the solution in the schematic I attached? When you switch of the reverb, the output of the OD relay goes directly to the master volume. The only thing this idea doesn't account for is the difference between lead dress between a stock ODS and with an added reverb. Note there are two traces: one with reverb, one without (bypassed). Can you see which is which by the trace alone?
If we are looking at the same schematic (Larry's) I don't see a separate trace - one with and one without reverb. In your simulation screenshot however, I do see a separate trace, from U101 to the PI in the bottom schematic without reverb. Whereas in the top simulation (and the middle) the master appears to be after the mixers coupling cap... U64 top, U65 middle.
Anyway, I have an alternative switching arrangement here:
ODS RVB 2.9.pdf
This bypasses the reverb and goes directly from the OD relay > RVB relay > Master > FX loop > PI and switching on the reverb it goes OD relay > RVB relay > RVB > Master > FX loop > PI. The only difference here is that the static line between the input of the RVB circuit and the master after the output is broken by the RVB relay... perhaps that was the original mistake that needed to be fixed?!
Not quite clear which sim screenshot you mean. The one I added a few posts ago? That is there just to bypass the reverb. IIRC I never used a bypass in any of my other designs and sims, let alone a trace hard bypassing the reverb.
Your design is better now. One serious word of precaution: you don't switch the output of the mixing amp (V9b). In this case, when you bypass the reverb, the .02 uF (C4) and output impedance of V9b will have a big impact on the sound and act like a low pass filter. I don't think you want that.
A few posts back I proposed a solution, but the attachment does not show. Let's hope it does now.
You do not have the required permissions to view the files attached to this post.
Not quite clear which sim screenshot you mean. The one I added a few posts ago? That is there just to bypass the reverb. IIRC I never used a bypass in any of my other designs and sims, let alone a trace hard bypassing the reverb.
Your design is better now. One serious word of precaution: you don't switch the output of the mixing amp (V9b). In this case, when you bypass the reverb, the .02 uF (C4) and output impedance of V9b will have a big impact on the sound and act like a low pass filter. I don't think you want that.
A few posts back I proposed a solution, but the attachment does not show. Let's hope it does now.
The sim screenshot I was referring to was the 124/005 hybrid you made... This present screenshot, is new to me, and I feel I'm getting somewhere now. I've run out of numbers for version 2 (!) so here is the next iteration, but this time with your bypass switching, taking the reverb entirely out of the circuit and circumventing that clanger of a low pass filter.
ODS RVB 3.0.pdf
I like this a lot more! What I am curious about is if it would be wise to place an arc-suppression RC net across the COM1 and no1 contacts of the relay (indicated by the red area in the schematic) If the AC current is fairly low, in the mA range, perhaps it wouldn't be necessary. If so, I understand the components would need to go on the relay board as close to the contacts as possible.
I see you have reduced the value of the grid leak resistor from 220k to 100k in your present simulation and removed the grid stopper but the mixer also has a smaller LNFB 10Meg resistor with an added 1Meg to ground. A different cathode cap value on the recovery and mixer too. How would you characterize these changes if you don't mind me asking?
It is going to take some time before I can read and inwardly digest Aiken's whitepaper on the mixer but I thank you for that. It's important for me that I don't just copy and paste someone else's work (as brilliant as it might be) but that I can understand it to the level where I can confidently shape the tone using my own insights. The maths doesn't bother me. Not knowing stuff, bothers me
You do not have the required permissions to view the files attached to this post.
Tony, is there much compression when playing softly
Here is how I would describe it..There is always some amount of compression as well as a slower attack going on at any volumes,vs the standard LTP. As you increase signal either on the input or master or on guitar the attack time slows down. (nuance) as you begin to turn the master up the amp gets louder and continues stays clean. In a normal type LTP PI. where you hit the onset of output tube distortion does not happen? instead the amp compresses further and the top and bottom end flattens out the louder the amp gets
could there be something going on with respect to GNFB in an amp that is either A.C or D.C Coupled we could be missing here?
As Rootz points out, the CF's in 002's mixer stages are DC coupled to the voltage amplifier stages feeding them, which is the situation described by Blencowe. The DC coupled CF power stage driver in SSS 002 is the opposite, DC coupled to the output tube grids and AC coupled to the PI. The driver stage bias point is at around -4V, so nowhere near any continuous grid current flow.
AFAIK All SSS's are AC driven (Cap before the PI) my point was simply if Rootz was wondering if this compression effect happens due to the current starve the followers have in the Tweed mixer that may not be the case since 004 and 005 do not use the followers and still have this same type compression effect AFAIK
BTW. Just so we are clear on what I refer to as A.C coupled and D.C coupled (in the context of Dumble's use of P.I Drivers)..AC uses coupling caps on the way to the grids (Winterland/Odyssey) and D.C has no coupling caps (SSS)
Tony
Last edited by talbany on Sun Apr 11, 2021 9:36 pm, edited 2 times in total.
" The psychics on my bench is the same as Dumble'"
DC coupled: there is no coupling cap in series with the CF grid. AC coupled: there is a coupling cap in series with the CF grid. So in 002 the output tubes are DC coupled, but the CF drivers before them are not. Anything I'm missing on this matter? Martin?
Yes and further in the configuration Blencowe refers to as "current stealing," the CF grid is connected directly to the plate of the previous stage, as it is in the mixer section of 002 and the stage driving the tone stack in the Tweed Bassman.
That second schematic is not one you should use: many mistakes; configurations that simply won't work. However, it does show the difference between AC and DC coupled CF's.
Here is a plan of the chassis over the top of the layout, showing the two extra reverb tubes.
It's an Acrobat file in layers again, click the layers icon to view and select.
Dumble chassis - 9 tubes.pdf
This is a standard Dumble ODS chassis as far as I can tell. There isn't much room left and if any more tubes were to go in, it would take a lot of changes. As it is, the output transformer would need to go a little to the left, in this view, though there may be some play with the lead dress that won't impact its position in relation to the power transformer and the choke. The choke requires careful positioning if I read it right. My one concern here, is to keep the tubes far enough apart that they don't suffer EM interference. Does anybody have any advice for placing the reverb transformer? I've seen it placed diagnonally in relation to the chassis - why is that? and are there any rules of thumb for finding the optimum position?
Good health!
Stephen
You do not have the required permissions to view the files attached to this post.
talbany wrote: ↑Sun Apr 11, 2021 9:45 pm ... Here is the Pentode amp ...
Off topic here, therefore only briefly to avoid misunderstandings: As far as I still remember the Pentode amp correctly after now around 25 years, the schematic talbany posted here isn't a correct representation of its circuit. E. G.: AFAIR V2 in the real amp is wired completely different than V2 in the schematic talbany posted here. So the designation of this schematic as “Inspired by Dumbleland Silver Alligator” seems to be chosen appropriately.
