Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: kagliostro on January 02, 2020, 03:19:23 am
-
Hello friends have an Happy New Year
---
When an amp is arranged like this
(https://i.imgur.com/HeCaqst.jpg)
There isn't a resitor on the input of V2 pin # 2 (like R2 on V1 pin #1)
I was wondering if it will be better to add a resistor in that position
If I add it the radio interferences are avoided when the second input is used, but signal is reduced ....
so ... what to do ?
Add a resistor and modify the value of the resistors on the preceding partition resistor ?
Thanks
Franco
-
A grid stopper as per R2 doesn’t reduce signal level within the normal frequency bandwidth.
So yes, one should be added.
As they’re in series with the signal, any hiss from them will be added to the signal. Such noise is proportionate to resistance value, and if too high, due to Miller capacitance, high frequencies will be rolled off. Hence we don’t tend to see values over 68k in tube guitar amps (typically two 68k resistors in parallel are used, so an effective grid stopper of 34k).
The minimum value for a grid stopper to be useful is 7/gm, so to accommodate tube ageing, at least 6k8 for a 12AX7.
See https://www.aikenamps.com/index.php/grid-resistors-why-are-they-used
https://www.aikenamps.com/index.php/what-is-miller-capacitance
FYI to avoid the signal to noise ratio being avoidably degraded due to hiss and hum noise sources being introduced, input stage cathodes should fully bypassed.
-
input stage cathodes should fully bypassed
don't have my notes, seems full bypass in guitar range was ~~ 22uf, but in guitar world even a tone deff'r like me notices the "muddying" up at 22uf, I agree some cap is warranted though
-
(http://el34world.com/Forum/blob:https://imgur.com/5155abd9-65d2-41a3-8aa3-6c38908e1784)Thanks Pdf64 & Shooter
OK I understand
---
If I add a 33K (or, at least, a 10K resistor), considering that in that amp there is a partition resistor that put a 680K resistor in series with the signal and 220K connected to ground, do you think that adding a 33K resistor it will be better to modify the partition resistor value to compensate (620K + 30K = 650K) + (200K + 10K =210K) ?
--
Meanwhile I remembered I've some particular input female jack with switchs like this
(https://i.imgur.com/8kDQidB.jpg)
(but I don't remember exactly how the switchs acts, I must find it and control)
Do you think that exist an input jack with switchs that will permit to have the added in series resistor on the path of the signal only when the jack is plugged in ?
Thanks
Franco
-
(http://el34world.com/Forum/blob:https://imgur.com/5155abd9-65d2-41a3-8aa3-6c38908e1784)Thanks Pdf64 & Shooter
OK I understand
---
If I add a 33K (or, at least, a 10K resistor), considering that in that amp there is a partition resistor that put a 680K resistor in series with the signal and 220K connected to ground, do you think that adding a 33K resistor it will be better to modify the partition resistor value to compensate (620K + 30K = 650K) + (200K + 10K =210K) ?...
I don't think you seem to understand, sorry.
The grid stopper in no way affects the potential divider, ie the signal level at the 2nd stage grid doesn't change whether there's a grid stopper on it or not, over the audio frequency band we're concerned with.
input stage cathodes should fully bypassed
don't have my notes, seems full bypass in guitar range was ~~ 22uf, but in guitar world even a tone deff'r like me notices the "muddying" up at 22uf, I agree some cap is warranted though
To minimise the cathode picking up hum from its heater, the cathode's 'full bypass' must extend down to somewhat below mains frequency.
In regard of an input stage (ie which generally operate within their linear range), dunno how that could cause the sound to get muddy? For later stages then yes, they will be liable to being overdriven and hence larger bypass cap values will affect the recovery time.
Here's a peer reviewed cathode bypass calculator that uses the actual equation provided in RDH4 http://bmamps.com/CapCal.html
-
Thanks Pdf64
Surely I explained badly what I was thinking
I was meaning to modify the value of the partition resistor not to compensate the loss in signal, I was thinking to avoid the Hiss due to the added resistor that increases the total resistance on the signal path
original 680K, adding a 33K resistor it will be 680K + 33K = 713K
Franco
-
I don't think it would work like that :w2:
By 'partition resistor', I think you may mean 'potential divider' resistor?
-
dunno how that could cause the sound to get muddy?
no argument the calculations are correct, my observations/understanding, too much low end causes downstream mud, so by limiting the low end at V1A, you keep unwanted interference limited, and satisfy the guitar folk, win win :laugh:
side observations, guitar folk can and do enter V1A with enough distortion already that keeping gain and SNR lower at the low end helps tweaking in the downstream
-
@pdf64: In regard of an input stage (ie which generally operate within their linear range), dunno how that could cause the sound to get muddy? For later stages then yes, they will be liable to being overdriven and hence larger bypass cap values will affect the recovery time.
I routinely use 1uF bypass caps on the 1st stage. Sounds better to me, and many others. 22uF "boosts" the entire audio range or more. For guitar tuned to pitch any AC in the signal below 80 Hz might be considered "mud"; which gets re-amplified in subsequent gain stages. It is also a waste of amplifier energy.
@kagliostro: Per your schematic, the grid stopper R is NOT part of a voltage divider. OTOH, if the grid stopper were to the Left (input side) of the grid leak R, then there is a voltage divider (though the amount of voltage reduction would be small). But, for best performance as a grid stopper, it should soldered with short lead directly to the tube lug.
-
I was meaning to modify the value of the partition resistor not to compensate the loss in signal, I was thinking to avoid the Hiss due to the added resistor that increases the total resistance on the signal path
original 680K, adding a 33K resistor it will be 680K + 33K = 713K
The total noise is not the algebraic sum of the two noise sources. The equation for the total noise involves adding the squares of each noise voltage and then taking the square root of that sum. This means that if one noise voltage is somewhat larger than the other, then the total noise voltage is essentially the same as the larger noise voltage by itself. The smaller noise voltage effectively drops out of the equation.
For example, let's say that R2 (33K) generates 3nV of noise and it gets multiplied by 50 by the first gain stage. The voltage divider between stages knocks the noise back down to 31.5nV. Now let's say that 8nV of noise is generated between the first and second gain stages. The two noise voltages don't simply add to get 39.5nV, but instead the square root of the sum of the squares equals 32.5nV. Even though the noise generated between the first and second stages was much larger than that generated by R2, the total noise barely changed.
-
The goal of removing low frequency 'mud' would seem to be rather more effectively achieved by the use of a high pass filter after the 1st stage, rather than a partial bypass of the 1st stage cathode?
As a full bypass of the 1st stage cathode, followed by a high pass filter, has the double whammy of minimising hum packed up by the cathode, and then filtering out any that does get picked up, along with the LF mud.
Whereas a partial bypass is a shelving filter, hence inherently less effective than a high pass filter at removing mains freq interference, all else being equal.
It seems poor engineering to intentionally compromise the input stage's signal to noise ratio to try to achieve a goal that can be better met by other means; much better to use the right tools for the job.
-
It seems poor engineering
:l2:
the 1st guitar amp I put on my bench horrified me as I scoped out the good signal path
complex harmonic content designed basically by musician's ears is a complete other field of engineering :icon_biggrin:
-
Much of this thread is theory unconnected to practice. My ViboChamp, as modded, is dead quiet. I guess I lucked-out. It was so quiet, I turned the controls all the way up & still heard nothing; assumed it wasn't functioning, struck a guitar chord and darn near blew my eardrums out. This with twisted heater wiring; 2X 1uF bypass caps, 3X 20uF filter caps; and a 6L6.
kagliostro reports a problem with noise. You could just as well approach that from some other direction than the cathode bypass cap. Some types of noise are subject to death by grid stopper, e.g. In my anecdotal experience, large K bypass & B+ filter caps are not necessary for a quiet amp.
-
kagliostro reports a problem with noise.
I don't think he is reporting anything. I believe he is just wondering why and what if. :icon_biggrin:
-
and we've contributed ALOT of what if's n some pretty good Y's :icon_biggrin:
I like mined pushups way better than the arm ones :laugh:
-
kagliostro reports a problem with noise.
I don't think he is reporting anything. I believe he is just wondering why and what if. :icon_biggrin:
I thought he was actually having radio interference. :dontknow:
-
Many Thanks to ALL
Ciao Jjasilli
I don't think he is reporting anything. I believe he is just wondering why and what if.
Steve is right I'm just wondering
Franco
-
...My ViboChamp, as modded, is dead quiet. I guess I lucked-out. It was so quiet, I turned the controls all the way up & still heard nothing; assumed it wasn't functioning, struck a guitar chord and darn near blew my eardrums out. This with twisted heater wiring; 2X 1uF bypass caps...
Yes, I think there's either good luck, or making sure to only buy 12AX7 that are checked for low hum (ie low h-k leakage).
Whatever, it's nice to use good practice to make an amp's hum level more resilient to bad luck / non-ideal tubes.
As mentioned, cathode bypass is a shelving filter; if that gives the desired response, it would be fairly trivial to replicate that response with a full bypass of the input stage cathode and implementing a shelving filter between the stages.
Dunno why that approach would attract debate :w2:
Whatever, I'm just suggesting it as being what seems to me an obvious example of good practice, I'm perplexed by the derision and controversy it's attracted, especially given the nature of the OP.
-
I'm perplexed by the derision and controversy it's attracted,
no derision here, my career was understanding how n why to fix it, not create the "next gen", so I come at it from what 50yrs of real(ish) engineers did specifically in guitar tube amps, stereo tube amps have a whole "other" set of observations and norms. At the end of day though, my goal is to separate musicians from thier money with a quality product, an audiophile would tell me to pack sand with the same amp.
want to go completely wankers, talk to a guitar player about MOJO and try to convert that to engineering :icon_biggrin: