I need a gainstage after Nadja (active xover)for the basschannels,but I dont need much voltage about 1-2vRMS.
So I am studing the data sheets and it has a gain of 20,but a bit further down
https://www.google.se/url?sa=t&rct=...N7.pdf&usg=AFQjCNEDY86xmAf_ULZaOgd0_mtdv76MsQ
I see they are running a B+ of 90v! for a again of 14,will this work for the basschannels?
I can adjust nadja to give a maximum 1v out.And lower.
AFAIK this is the smallsignaltube with the lowest gain,or is there one even lower?
So I am studing the data sheets and it has a gain of 20,but a bit further down
https://www.google.se/url?sa=t&rct=...N7.pdf&usg=AFQjCNEDY86xmAf_ULZaOgd0_mtdv76MsQ
I see they are running a B+ of 90v! for a again of 14,will this work for the basschannels?
I can adjust nadja to give a maximum 1v out.And lower.
AFAIK this is the smallsignaltube with the lowest gain,or is there one even lower?
If you want gain of 2X ~ 4X, then an anode follower circuit may be your best option. This page provides a good introduction: The Valve Wizard
Not really a small signal tube but 6BX7 is pin compatible and sounds fantastic in preamps. Heater current a little high perhaps at 1.5amps. I have a pre that takes 6AS7 for lowest gain, then 6BX7, 6BL7, and finally 6SN7 which gives most gain. They all work great even tho none are optimized in this circuit.
If the cathode is bypassed with a cap, removing it will lower the gain. Secondly, although most people will baulk at this idea, you can obtain a lower gain by using a resistor divider between the input cap to the grid. Think of a pot but with fixed values. eg. if you want to cut gain by half connect two 50k resistors in series, tie one end to the input cap, the other end to grond, and take the grid signal from the centre point.
Cheers.
Cheers.
Excellent point! But for some people, open loop amplification is preferred Personally I feel there is a time and place for both.
Cheers.
That depends on other factors associated with whichever kind of feedback is used.
I myself would rather use an unbypassed cathode resistor if the increased rp can be tolerated. (It might mean biasing the grid from a tap on a large cathode resistance for enought effect.) Otherwise series feedback (anode - grid), but then the grid series resistance would add noise if large and depending on the signal level. Lowering the h.t. voltage to get lower gain would not be on my list; getting to low enough with a low-µ tube to begin with to have a worth-while effect, brings the operation into a non-linear region operating the tube outside its intended field.
Or as Kobamix said, in the final instance a voltage divider, at the input or output depending on where in the signal line the stage is used.
I myself would rather use an unbypassed cathode resistor if the increased rp can be tolerated. (It might mean biasing the grid from a tap on a large cathode resistance for enought effect.) Otherwise series feedback (anode - grid), but then the grid series resistance would add noise if large and depending on the signal level. Lowering the h.t. voltage to get lower gain would not be on my list; getting to low enough with a low-µ tube to begin with to have a worth-while effect, brings the operation into a non-linear region operating the tube outside its intended field.
Or as Kobamix said, in the final instance a voltage divider, at the input or output depending on where in the signal line the stage is used.
Indeed, as inferred, SemperFi;
But to add figures (and hopefully not to bore):
Referring to the given diagram, the tube feedback resistor Rf there is effectively in parallel with Ra. Thus so as not to load down Ra, Rf needs to be quite higher than Ra, in turn 'forcing' the series input grid resistor Rg to often be several 100K for a low gain. A noise problem with the series resistor: For audio the resistor noise can be of the order of 2 - 4µV per 100K, which may not sound like a lot but as said, for a low signal level input might begin to count. That small a signal though does not appear to be what the OP wants to use it for, so: Indeed!
(For a 12AT7/6SN7 e.g. Ra = 47K; Rf = 180K; Rg = 82K for gain of the order of 2.)
But to add figures (and hopefully not to bore):
Referring to the given diagram, the tube feedback resistor Rf there is effectively in parallel with Ra. Thus so as not to load down Ra, Rf needs to be quite higher than Ra, in turn 'forcing' the series input grid resistor Rg to often be several 100K for a low gain. A noise problem with the series resistor: For audio the resistor noise can be of the order of 2 - 4µV per 100K, which may not sound like a lot but as said, for a low signal level input might begin to count. That small a signal though does not appear to be what the OP wants to use it for, so: Indeed!
(For a 12AT7/6SN7 e.g. Ra = 47K; Rf = 180K; Rg = 82K for gain of the order of 2.)
> 6SN7 low plate voltage for lower gain? ... it has a gain of 20.... B+ of 90v! for a again of 14...
Look again.
The "20" is an ideal value, can not be hit with any simple circuit.
The gain of a tube stage *hardly* changes with supply voltage. Look at the top line in my attachment. Gain is 14 for B+ of 90V, 180V, or 300V. Even with far higher loading values, gain only changes from 14 to 16, hardly over 1dB thus hardly audible.
What low B+ does is give less output for the same distortion or more distortion for the same output.
If you want less gain:
* Leave out the cathode bypass, as kodabmx says. This typically cuts gain in half, also reduces THD, but does raise output impedance some.
* Put a voltage divider in front, kodabmx's other suggestion. This gives a VERY simple way to adjust to suit your system.
* Put NFB around it. Mike's link to Merlin's page shows the basics. Distortion and output impedance will drop. Input impedance typically falls, but you can usually keep this in reason.
> use high mu ... negative feedback,or can it be used on 6SN7 as well?
The question is: how much gain are you throwing-away? For same excess-gain, the improvements are generally the same. Take 12AX7 at full gain of 50. NFB for gain of 16. 3:1 thrown-away. Now ponder 6SN7/12AU7 with full gain of 14, NFB for gain of 4.6, same 3:1 throw-away. Both will be about 3:1 down on their full THD, and both will be similar THD for same supply voltage and output level. The 12AX7 job will have Zout near 13K and the 6SN7 will be under 3K Zout. (If both were taken down to gain of 4.6, the Zouts would be similar and the 12AX7 may be a shade lower THD.)
Look again.
The "20" is an ideal value, can not be hit with any simple circuit.
The gain of a tube stage *hardly* changes with supply voltage. Look at the top line in my attachment. Gain is 14 for B+ of 90V, 180V, or 300V. Even with far higher loading values, gain only changes from 14 to 16, hardly over 1dB thus hardly audible.
What low B+ does is give less output for the same distortion or more distortion for the same output.
If you want less gain:
* Leave out the cathode bypass, as kodabmx says. This typically cuts gain in half, also reduces THD, but does raise output impedance some.
* Put a voltage divider in front, kodabmx's other suggestion. This gives a VERY simple way to adjust to suit your system.
* Put NFB around it. Mike's link to Merlin's page shows the basics. Distortion and output impedance will drop. Input impedance typically falls, but you can usually keep this in reason.
> use high mu ... negative feedback,or can it be used on 6SN7 as well?
The question is: how much gain are you throwing-away? For same excess-gain, the improvements are generally the same. Take 12AX7 at full gain of 50. NFB for gain of 16. 3:1 thrown-away. Now ponder 6SN7/12AU7 with full gain of 14, NFB for gain of 4.6, same 3:1 throw-away. Both will be about 3:1 down on their full THD, and both will be similar THD for same supply voltage and output level. The 12AX7 job will have Zout near 13K and the 6SN7 will be under 3K Zout. (If both were taken down to gain of 4.6, the Zouts would be similar and the 12AX7 may be a shade lower THD.)
Attachments
Preference for no feedback requires usage of highly linear tubes in highly linear regimes. It does not mean that "A tube that looks nice to me" will not sound like a merde without feedback.
Preference for no feedback requires usage of highly linear tubes in highly linear regimes. It does not mean that "A tube that looks nice to me" will not sound like a merde without feedback.
Or a tolerance and indeed admiration of the slight non linearity of the humble triode.
I've built a triode connected 6L6 parallel push pull amp that sounds great with no nfb. It even sounds good in tetrode mode. I do use it to reduce gain when running tetrode but the linearity of a 6sn7, 6dj8, 6n1p, 6n3p, 6n5p, 6n6p, 6n8s etc are plenty linear for my ears. Otherwise I'd use op amps.
I had to use 2 of 6P3S-E in parallel in a PP amp, in class A, with nested feedbacks, to get clean 40W per channel. "Clean" I mean, a real high end, when there is an illusion as if no amps, no speakers, just talents here and now.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.