I've made a bass preamp with two 5670's. It has two channels with #1 being the "simple" one - only a volume control - and #2 being the "complicated" one with the fender twin style TMB tone stack. Here's the schem:
Voltages measured:
B+ = 286V
anodes: 112V to 122V
cathodes: 3.3V to 3.58V
when channels run in parallel everything's very good and you can only wonder how much gain TMB runis - channel one is at least twice louder.
with channels in series (channel 1 out into channel 2 in) oscillations start when there's too much volume on either channel. Like, with c.2 volume on maximum everything sounds good untill volume on channel 1 is under 3 (about 450K on the 500K pot). When you pass 3 and the vol. pot resistance is around 400K a nice squeal starts and if you go further you get a nasty oscillator scream. If you max the channel one and lower the second channel's volume there's about the same thing, only you get more distortion before the squeal starts and it also happens when the volume (tihs time on ch.2) is less than 400K.
Bass preamp should be more in a hi-fi preamp field (it's NOT a guitar pre), I don't need any distortion at all. I was thinking about some solutions but cannot deside which one is the best.
My primitive ideas include: puting a series 220K resistor on channel 1 output; lowering the ground reference resistors on both channels outputs to, say, 100K.
Or maybe everything is wrong and the tubes should be biased differently to get rid of the oscillations?
by the way, I like the sound of the preamp, although there's something wrong when the channels are in series close to the oscillation point - some additional harmonics appear when a strong signal is on the input. What's that?
Voltages measured:
B+ = 286V
anodes: 112V to 122V
cathodes: 3.3V to 3.58V
when channels run in parallel everything's very good and you can only wonder how much gain TMB runis - channel one is at least twice louder.
with channels in series (channel 1 out into channel 2 in) oscillations start when there's too much volume on either channel. Like, with c.2 volume on maximum everything sounds good untill volume on channel 1 is under 3 (about 450K on the 500K pot). When you pass 3 and the vol. pot resistance is around 400K a nice squeal starts and if you go further you get a nasty oscillator scream. If you max the channel one and lower the second channel's volume there's about the same thing, only you get more distortion before the squeal starts and it also happens when the volume (tihs time on ch.2) is less than 400K.
Bass preamp should be more in a hi-fi preamp field (it's NOT a guitar pre), I don't need any distortion at all. I was thinking about some solutions but cannot deside which one is the best.
My primitive ideas include: puting a series 220K resistor on channel 1 output; lowering the ground reference resistors on both channels outputs to, say, 100K.
Or maybe everything is wrong and the tubes should be biased differently to get rid of the oscillations?
by the way, I like the sound of the preamp, although there's something wrong when the channels are in series close to the oscillation point - some additional harmonics appear when a strong signal is on the input. What's that?
Nice job!
With more than 2 cascaded stages, decoupling is a must. Feed the B+ to each stage with a 10K resistor and 22 uF or larger bypass - this will keep anything above 5 Hz from sneaking from stage to stage through the power supply. But 4 stages is just TOO much gain - you have a potential gain of 30*30*30*30, with input and output 1/4" apart on that switch! Removing cathode bypass caps might be a start...
With more than 2 cascaded stages, decoupling is a must. Feed the B+ to each stage with a 10K resistor and 22 uF or larger bypass - this will keep anything above 5 Hz from sneaking from stage to stage through the power supply. But 4 stages is just TOO much gain - you have a potential gain of 30*30*30*30, with input and output 1/4" apart on that switch! Removing cathode bypass caps might be a start...
Feedback through the power supply, at the least shunt that last electrolytic with a good film cap to lower its impedance at high frequencies, better still split off one stage before and power each gain block from its own decoupled supply. Keep voltages about the same.
Add grid stopper on second stage 5670 right at the socket.
Unbypassed cathode resistors will add some noise particularly at high gains. I would instead add an output attenuator for use when you loop things in series - this would ideally be in the first stage output and the attenuated output would be used to drive the second stage with the tone stack. I don't think you really want all of that gain anway..
Add grid stopper on second stage 5670 right at the socket.
Unbypassed cathode resistors will add some noise particularly at high gains. I would instead add an output attenuator for use when you loop things in series - this would ideally be in the first stage output and the attenuated output would be used to drive the second stage with the tone stack. I don't think you really want all of that gain anway..
options
I think I simply need to get rid of too much gain. The oscillations start after a pretty heavy distortion which I don't need. I'm not sure where they bleed from but I'm afraid anything I will do will not help because it's just too much gain.
I would not add more grid stoppers after volumes just because it works good in parallel and any additional stuff in the signal chain might make it muddy.
An output attenuator at the end of the 1st stage seems to me to be the best option, it may be even good for the parallel since the first stage is much louder than the 2nd.
But what sort of attenuator? Voltage divider? Series resistor? low ground reference?
I think I simply need to get rid of too much gain. The oscillations start after a pretty heavy distortion which I don't need. I'm not sure where they bleed from but I'm afraid anything I will do will not help because it's just too much gain.
I would not add more grid stoppers after volumes just because it works good in parallel and any additional stuff in the signal chain might make it muddy.
An output attenuator at the end of the 1st stage seems to me to be the best option, it may be even good for the parallel since the first stage is much louder than the 2nd.
But what sort of attenuator? Voltage divider? Series resistor? low ground reference?
Stick the attenuation in series with the volume pot or because of the pots high value connect it across the lower resistor in the attenuator pair. You could also change to a lower value pot. My concern is high frequency performance could be degraded by the shunt capacitance across the pot. Tweak values so that levels match fairly well or do what you need.
The cross talk is in the supply, and I am doubtful that the electroytics you have chosen have particularly good performance above a couple of kHz. I'd still shunt the last one in the filter with a good film cap of a couple of uF.
The grid stoppers will not affect tone or frequency response, particularly in comparison to the source impedance that pot represents around mid point. I would recommend something on the order of 1K for much better resistance to rf pick up.
Kevin
The cross talk is in the supply, and I am doubtful that the electroytics you have chosen have particularly good performance above a couple of kHz. I'd still shunt the last one in the filter with a good film cap of a couple of uF.
The grid stoppers will not affect tone or frequency response, particularly in comparison to the source impedance that pot represents around mid point. I would recommend something on the order of 1K for much better resistance to rf pick up.
Kevin
chassis
yep, that's a security box made of steel. you can see the lock on the picture with the insides. I've torn the cover away but basically I could have left it and close the box with the key. It costs around $25 new and you also have biger sizes for around $30-35. I've made bigger amps with different boxes and they're very cool and robust.
I've taken this device to the rehearsal tonight. The problem seems to be pretty obvious - too much gain. Plugged into 300W power amp (Fender solid state) it was already too loud with 1/4 voulme in parallel mode. It won't even get close to the oscillation zone without blowing the speakers.
The preamp vas pretty quiet (it terms of noise of course) and even knocking on the tube's glass produced only minor clicks. 5670's are super non-microphonic. I still wonder how people LOVE those 12AX7's when there are much better tubes for much lower prices...
yep, that's a security box made of steel. you can see the lock on the picture with the insides. I've torn the cover away but basically I could have left it and close the box with the key. It costs around $25 new and you also have biger sizes for around $30-35. I've made bigger amps with different boxes and they're very cool and robust.
I've taken this device to the rehearsal tonight. The problem seems to be pretty obvious - too much gain. Plugged into 300W power amp (Fender solid state) it was already too loud with 1/4 voulme in parallel mode. It won't even get close to the oscillation zone without blowing the speakers.
The preamp vas pretty quiet (it terms of noise of course) and even knocking on the tube's glass produced only minor clicks. 5670's are super non-microphonic. I still wonder how people LOVE those 12AX7's when there are much better tubes for much lower prices...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.