• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Improving preamp's bass response.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Disabled Account
Joined 2004
I did make some effort to content the engineering side of this thread -me posting graphs? you got to be kidding me!- but to no avail. Boomy responses was all I got. I'm satisfied with the end result though, did learn a few things and, in general, this thread has been fun to read. My preamp doesn't have a better bass because of it but this doesn't trouble me. Maybe I should repeat again: I don't have a problem with my bass, just not the best one around and that is all.
 
Well, someone quoted a 6.3 Hz cutoff frequency from a 47 uF cap. And another mention was made of a 5 Hz cutoff frequency. Those -3 dB frequencies do seem rather high, if you're worried at all about bass response.

Usually, you would want a -3 dB cutoff frequency that is no more than 1/10th (or 1/100th for purists) of the lowest frequency that you want to remain unaffected.

So maybe you should triple those capacitances that were mentioned, to get the cutoff frequencies down to 2 Hz or less, so your bass might be decent down to 20 Hz. And maybe it would be quick and easy to test by just temporarily clamping a 100 uF cap in parallel with the 47 uF, and so on (but do them all at once, not one at a time).

I didn't look at the circuit (and am not a tube circuit expert, anyway) so hopefully someone will chime in if any of those changes would also require something else to be changed.
 
I am not sure where you get your information, but that is untrue for a line stage amp. The difference in frequency response between vacuum tubes and solid state will be indistinguishable.


What exactly is untrue? That there is an audible difference?

Frequency response says very little about anything, especially for a line stage where obtaining a level response is trivial. Btw, most solid state preamps are dc coupled, so on top of differences in active devices, tube amps usually have coupling caps with their specific sonic contributions.
 
What exactly is untrue? That there is an audible difference?

Frequency response says very little about anything, especially for a line stage where obtaining a level response is trivial. Btw, most solid state preamps are dc coupled, so on top of differences in active devices, tube amps usually have coupling caps with their specific sonic contributions.

Well, first, two competently designed line stages, regardless of topology (solid state or valve), should be indistinguishable over the audio spectrum.

Yes, DC coupled circuits can pass 1 Hz signals, but 1 Hz or even 10 Hz does not impact audible bass for music. The lowest a piano goes is 27.5 Hz and you won't hear anything below about 20 Hz anyway.

Second, there is no magic in solid state that artificially boosts bass any more than there is a deficiency in vacuum tubes that suck bass out. Again, a poorly designed line stage can do that and it can be done regardless of topology.

Third, any distortion injected by components such as capacitors is just that; distortion. It is not going to suck out bass - unless it is incorrectly sized for the job. However, I already stated that we are comparing competently designed circuits, not something designed correctly to something designed incorrectly. That comparison, for the sake of arguing topology, is useless.

However, if you have some scholarly evidence that supports your claim that vacuum tubes can not pass audible bass frequencies as well as solid state, I would be happy to review it.
 
Well, someone quoted a 6.3 Hz cutoff frequency from a 47 uF cap. And another mention was made of a 5 Hz cutoff frequency. Those -3 dB frequencies do seem rather high, if you're worried at all about bass response.

Usually, you would want a -3 dB cutoff frequency that is no more than 1/10th (or 1/100th for purists) of the lowest frequency that you want to remain unaffected.

So maybe you should triple those capacitances that were mentioned, to get the cutoff frequencies down to 2 Hz or less, so your bass might be decent down to 20 Hz. And maybe it would be quick and easy to test by just temporarily clamping a 100 uF cap in parallel with the 47 uF, and so on (but do them all at once, not one at a time).

I didn't look at the circuit (and am not a tube circuit expert, anyway) so hopefully someone will chime in if any of those changes would also require something else to be changed.

Cassiel,

This would be so easy to test. Did you lose interest?

Tom
 
This is a tube circuit configuration I've not seen before, but tracing out the circuit and especially knowing the values helps to show what's happening.
Anyway, even though I haven't had the time to try out possible fixes I have spotted a problem with one tube: one of the triode's filaments was underpowered, seriously underpowered. Barely glowing. The triode with problems was facing the wall so I really don't know when the problem started. Turns out my bakelite socket is acting up; my fault really, I have a soft spot for vintage parts and sometimes **** happens. Now I have to replace the sockets and possibly change the schematic. No more news from me in a while. Thanks to all.
If you must use "old" parts, go for high quality NOS (New Old Stock, unused parts). They might be more expensive, but I know I've got some ceramic miniature tube sockets somewhere that cost me something like a couple dollars each.
Value of R2 is 130K.
Great, now that explains how the circuit works. The tube is acting as a fairly high gain (maybe 20 to 100 - regardless, it's a lot higher than the whole circuit), with a negative feedback path through R2.

This is working like an op amp with the positive input grounded. The signal is going through R1 (100k) to the negative input, and the output is connected back to the negative input through R2 (130k), thus the gain is -1.3.

This circuit also makes the grid a virtual ground, so the input impedance seen by the instrument is 100k (or for an "exact" number, it's in parallel with 1M, so the impedance seen is 90.9k), significantly lower than is good for most electromagnetic pickup instruments. The instrument needs to see 1M, and the 1M to ground just provides an illusion. Due to the feedback resistor, the grid is effectively a very LOW impedance, not a very high impedance that it is in more conventional circuits. If this is meant to be some sort of tube buffer, it's doing a bad job. The effect of this load on the pickup must be taken into account to get an accurate frequency response.

Also, since (once again) the grid is a virtual ground, the impedance the output "Cap" drives is RL in parallel with R2 (56.5k), so the resulting low frequency cutoff should be recalculated (though I don't think that's a problem).

Even more, a bass guitar only goes down to 41Hz, unless it's a 5-string bass with a low B, which is 31Hz.


Musically, it may be this lower impedance loading the pickup that makes the mids magic you like, and it may not be easy to get the lows back the way a fully buffered circuit would without changing the mids. Maybe the lows could be boosted in a second stage added to the output of this one.
 
Disabled Account
Joined 2004
Cassiel,

This would be so easy to test. Did you lose interest?

Tom

No, I did change the circuit to SRPP and now it sounds x20 better than the SS preamp so who cares if the bass is a little softer. Anyway, the change has improve dynamics and there's a little more punch so I'm happy now. I have a small problem though, my speakers are VERY sensitive and now I have some high frequency noise. And when the volume pot goes to five there's some whistling... looks like the circuit is prone to oscillation; I'm going to install grid stoppers.

benb - thanks for the analysis. The magic of so much feedback is the low noise - there isn't any! And it sounded better than just having the 6SN7 in parallel with no feedback loop.
 
Or... how about a trivial solution?

Bourns 70F501AF-RC open ferrite 0.50 H inductor as the cathode bias? Resistor of 1K across it, for HF response. It'll emphasize bass, for sure. Series resistance of 735 ohms nominal, so at most there'll be a 5 dB bass boost, without resorting to capacitors. Ferrite open-core gives response up to MHz region. Compact. Cheap ($2.65 on Mouser) per unit cost. Can be used in conjunction with LED bias if that's desired.

Dunno. Sometimes all the theory is nice, but practicalities of L-R circuits make themselves known after some thinking.

GoatGuy
 
Thanks guys. Rk is 470 uF. One 6SN7 per channel (parallel triodes). I will start with the changes described here - bigger output cap and, also, that cap in the feedback loop and see what happens. I don't have the equipment to measure the frequency response, alas, I'm just a tinkerer - far, far away from being a pro.
months ago i experimented with my headamp, i was checking various ways how to connect double triodes.
started first on scope with paralel triodes with plate resistor. not happy, i liked single triode bit more.
but when compared to srpp, immediately liked it. srpp has balls to give hi output voltage (more than standard connection), and benefit of low out impedance.
(more output voltage, without early coming limitation)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.