Hoffman Amplifiers Tube Amplifier Forum

Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: kagliostro on November 11, 2014, 05:51:04 pm

Title: Is possible to arrange a variable gain stage ?
Post by: kagliostro on November 11, 2014, 05:51:04 pm
I mean if is possible to arrange a sensible variation on the gain of one stage

something like this

(http://i.imgur.com/u8RsqSM.jpg)


Thanks

K
Title: Re: Is possible to arrange a variable gain stage ?
Post by: HotBluePlates on November 11, 2014, 06:50:44 pm
Those would all probably work with varying degrees of effect.

Every place you have a switch, I'd probably use a 1MΩ or similar across the switch contacts so that the switch shorts the 1MΩ resistor when you want to "turn on" that particular circuit. That may help with reducing d.c. shifts and popping when you activate a switch.
Title: Re: Is possible to arrange a variable gain stage ?
Post by: kagliostro on November 11, 2014, 06:59:30 pm
You mean so ?

(http://i.imgur.com/7DR4omr.jpg)

Franco


Title: Re: Is possible to arrange a variable gain stage ?
Post by: terminalgs on November 11, 2014, 09:10:44 pm
Yes, that would work.  I bet the difference between  2.2uf and 1.2uf would not be noticeable in any way.  Pick one of those, and make the other 22uf (boosts all frequencies a guitar will kick out).

going from a 100K to 200K plate resistor will certainly change the total gain, but IMHO a little harder to tune to your preferences.  I'd go with the one or the other  I guess the 200K ? and pick a single cathode resistor and go with it. You can switch out a  couple of cathode bypass  caps to vary the frequency floor of what gets boosted if that's something you are after.  Then, if you want to switch between "high" and "low" gain,,  I'd switch out a resistor on top of your volume pot that follows (somewhere).

like., If your subsequent volume pot is a 1M,  put a switched resistor 470K (or 220K or whatever) in series with the signal and the volume pot ("on top" on the vol pot). easier to tune to your liking (swap out the 470K , 220K to whatever).
Title: Re: Is possible to arrange a variable gain stage ?
Post by: 2deaf on November 11, 2014, 09:44:50 pm
If you put a switch on the 5uf cap. in the left diagram, you could change the gain from 30 to 50 (or so) with typical voltages.  The middle diagram will increase the gain around 25% when you go from 100K to 200K, but your headroom will decrease.  Going from 3K to 1.5K will increase the gain less than 10% and the cathode cap. switch won't change the gain at all at most frequencies.
Title: Re: Is possible to arrange a variable gain stage ?
Post by: 2deaf on November 11, 2014, 09:53:23 pm
I like these strategies for changing gain because they leave the frequency response intact (if you keep the capacitor values up).  A 470K resistor in series with a 1M pot rolls off a lot of treble. 
Title: Re: Is possible to arrange a variable gain stage ?
Post by: kagliostro on November 12, 2014, 01:32:33 am
Many Thanks HotBluePlates - Terminalgs - 2deaf

Franco
Title: Re: Is possible to arrange a variable gain stage ?
Post by: tubenit on November 12, 2014, 05:27:40 am
Kevin O'Connors  Soma18 used a treble pot like this which also increases the gain.  BTW, there is VERY little insertion loss with this type of tone stack arrangement.

With respect, Tubenit
Title: Re: Is possible to arrange a variable gain stage ?
Post by: rzenc on November 12, 2014, 05:51:15 am
Kevin O'Connors  Soma18 used a treble pot like this which also increases the gain.  BTW, there is VERY little insertion loss with this type of tone stack arrangement.

With respect, Tubenit


I tried this type of control with great success on some hypergain amps. It helps to control treble content. I ended up with a fixed resistor later.




Switching cathode caps have a very dramatic effect on gain and frequency response.


When I first read the topic title, I thought you wanted a variable mu comp type inside the amp....


Best Regards


R.
Title: Re: Is possible to arrange a variable gain stage ?
Post by: kagliostro on November 12, 2014, 06:34:49 am
Thanks Tubenit and Rzenc

Tubenit, that is interesting

Rzenc

Quote
I thought you wanted a variable mu comp type inside the amp....

if you have some info or link about such a thing I'm always interested to read/learn something new

Thanks friends

Franco
Title: Re: Is possible to arrange a variable gain stage ?
Post by: terminalgs on November 12, 2014, 09:07:29 am

I like these strategies for changing gain because they leave the frequency response intact (if you keep the capacitor values up).  A 470K resistor in series with a 1M pot rolls off a lot of treble.

lots of ways to deal with that,  Marshall did it like this:
(http://i.imgur.com/7WFtemv.png)

parallel 470pf cap.   If you want a gain switch,  break that 470K into two resistors,, and parallel one resistor with a switch... whatever combo you want,, 120K+330K, 2x 220K, etc.:

(http://i.imgur.com/loJS8lP.jpg)

C2 is chosen based on both R4+R5+P1 and R4+P1 being accpetable in terms of low-freq. roll-off.

C3 gets those high freq. that you are concerned about after adding R4+R5 to the top of P1.

If you don't like R4+R5+P1>1M,  then make it =1M  with P1=500K, and R4=R5=220K.  Of course, you can select C2,C3,C4 so that any selection of R4,R5,P1 will result in your desired freq. response.
Title: Re: Is possible to arrange a variable gain stage ?
Post by: 2deaf on November 12, 2014, 09:51:40 am
That's not even close to what I call an intact freq. response.
Title: Re: Is possible to arrange a variable gain stage ?
Post by: terminalgs on November 12, 2014, 02:53:23 pm
That's not even close to what I call an intact freq. response.


I'm all ears!

BTW: Franco didn't include the 'what comes next' part of the design to indicate where the signal goes once it appears on the plate,  so I offered an example using a JCM800 first input stage and broke the 470k R into two parts.  If the JCM800's particular values are not what you have in mind, and this  particular example is way off base in those terms,  what do you do with the signal once it appears on the plate to get the freq. response you are looking for?

also, by "intact freq. response" what are you referring to? a flat freq. response?

Title: Re: Is possible to arrange a variable gain stage ?
Post by: 2deaf on November 12, 2014, 03:39:15 pm
Quote
BTW: Franco didn't include the 'what comes next' part of the design to indicate where the signal goes once it appears on the plate,  so I offered an example using a JCM800 first input stage and broke the 470k R into two parts.  If the JCM800's particular values are not what you have in mind, and this  particular example is way off base in those terms,  what do you do with the signal once it appears on the plate to get the freq. response you are looking for?

I was just pointing out that you can change the gain without changing the frequency response right there at that one tube.  Leaving the gain the same at the tube and changing the gain later with resistors and capacitors is going to make it real hard to keep the exact frequency response.  I don't really care because you can approximate it good enough for most uses and anyways, you frequently intend to change the response after a stage.   

Quote
by "intact freq. response" what are you referring to? a flat freq. response?

Any given response, flat or otherwise.
Title: Re: Is possible to arrange a variable gain stage ?
Post by: Refugee on November 16, 2014, 05:45:41 pm
At a guess you want remote control of gain without sending audio to the remote volume control panel.
Have a look at some TV tubes from the RF section. They used DC to control the gain of the received signal amplifier and some of these tubes ended up being used in tube tape recorders with automatic record level control. The DC feedback went to g2.
There is absolutely no reason for them not working in an audio circuit even though they have a maximum operating frequency that is well above what we need. It is just a maximum and nothing stops us using one at a lower frequency.
I have found one with an extra triode as an added bonus. You surely must be able to find one of these locally but I have added a link for you to see the description for search words.
http://r-type.org/pdfs/30c17.pdf (http://r-type.org/pdfs/30c17.pdf)