What is the difference I will see if any by choosing to run a 12ax7a at 100 volts rather than 250 volts DC? According to the tube manual, both still have an amplification factor of 100. I am confused at which voltage I should choose. I am also deriving my voltage from a 12 volt source in which I am using a 555 timer running a mosfet to get my B+. Any suggestions? I would like to use both half's of the tube. Can a 12ax7a overdrive the second half of itself on purpose. I am looking for both clean and overdrive by just using one 12ax7a. It will then go on to an external amplifier of quality or to a board for recording. I am interested in a small space to put it all in. I have already built the power supply portion and can output 250 volts at 1.8 ma or about 150-175 or so at 5.5 ma. I can get a tad bit more but trying to keep it cool as the transformer is not that big.
So which is it I should use 100 volts or 250 volts on the plate?
thanks
So which is it I should use 100 volts or 250 volts on the plate?
thanks
a difference in available output swing is the first issue, but really, you've grabbed the tiger by the wrong end - reach for the tail...
By this i mean, you design the voltage amp stages to be able to drive the output stage. So you need to know what the drive requirement of the output stage is first before any reasonable and useful statement can be made about voltage gain design
By this i mean, you design the voltage amp stages to be able to drive the output stage. So you need to know what the drive requirement of the output stage is first before any reasonable and useful statement can be made about voltage gain design
thanks for the input, good points. The output is really only going to go to computer line level as that is what I have intended for it. So line level output is what I need. It is for a project I want to try out. I want to run the output to a computer audio input to record with it. If I cant overdrive the second half of the tube I will most likely be able to simulate it with computer program. I recall there is an overdrive simulator. So from what I gather from the post, for my purpose probably not a lot to worry, 100 or 250 volts for the swing, as I don't really have to drive anything. just output to, I think 10k load. I think 10k is the usual load for line level, I could be wrong. Is that correct then that 100 volts will do just fine?
Last edited:
Since this is guitar amplifier related I'm moving it to the Instrument & Amps forum per forum policy. It may also be of interest to the group there.
I'm not not that noob as it may seem. lol.. I will of course take steps to reduce that amount after I run it through the tubes and match it to the line level input of the computer. The whole idea behind this is to give the person the thrill of believing that they can get a tube sound since it went through a tube and using a name type of tube. Most of it is in the head above the shoulders in what they believe, of course. I have seen one in a magazine or online, dont have any details when or where though. Am I suppose to go to the musical part and post now and not here? I am a noob here as I am not here that often but would like to be as much as I can. I find all the info fascinating and rewarding. Besides the only place I know where I can actually talk electronics, nobody else to talk to. So let me know if I need to end this here and move to music section. Thanks.
Using a 12AX7 gives you two stages of gain. The first one will not get overdriven to any great extent by a guitar. It might be a good idea to have a FET as a gain stage before it to give the first tube something to work with. Maybe a gain of ten with a pot to adjust the amount of drive. While the tube can run on lower voltages it sounds better fed around the 250-300v range. You would want a pot to reduce the voltage going to the second stage also and then one at the output. Along the way you may want a toggle switch with three positions (on-off-on) with the center position with a 2.7k resistor on the cathode. Then you switch in a 3.3k which will give you 1.5k, or the other position with a 1.2k which will give you 830 ohms. This gives you a cold bias, center bias, and warm bias. They each have a different sound when overdriven. The cathode on the first stage should have a 0.68uF cap and a 22uF on the same type of toggle switch. This give an increas in gain but the 0.68uF cap limits the bass. Coupling caps of 0.22uF or less, 100k plate resistors. This should give you some variety. The 1.8mA is marginal but if it drops a few volts it should still be ok.
The 1st stage of a guitar amp needs 2 things, fairly clean amplification and low noise.
That low noise part dictates running at higher current (say 0.8mA to 1.0 mA) for a 12AX7. At that current you need plenty of anode volts such that the bias voltage required is above the -0.6V below which grid current becomes a problem.
You will definitely want the 250V rail rather than the 100V rail. For 100V you would want a different tube.
Cheers,
Ian
That low noise part dictates running at higher current (say 0.8mA to 1.0 mA) for a 12AX7. At that current you need plenty of anode volts such that the bias voltage required is above the -0.6V below which grid current becomes a problem.
You will definitely want the 250V rail rather than the 100V rail. For 100V you would want a different tube.
Cheers,
Ian
ok I am beginning to see the light. I would rather stay away from the FET. That would defeat the idea of the project. Not saying it is wrong but just wanting a tubes only circuit for ampification. If I need to beef up the power supply to add another 12ax7a I can. I have looked at other diagrams of guitar amps and about the cathode bias resistors; Instead of switches why not just say a 1k ohm resistor in series with a 5 k pot adjustable from the front panel? Or not. The 1 k to keep it from reaching zero at the low end of the pot. maybe a switch to switch caps in and out of circuit.
So I do need a clean circuit first and low noise?. Maybe that is all I need? Then let the computer simulate the other needs, such as overdrive etc. thanks for the ideas
So I do need a clean circuit first and low noise?. Maybe that is all I need? Then let the computer simulate the other needs, such as overdrive etc. thanks for the ideas
Printer2, I would love to see a rough sketch of a schematic of your FET idea.
As much as I consider myself a "tube nut" I can clearly see the benifits of running an FET in the first stage. Good clean stage with no requiered heater current, also microphonics would be less of a concern.
I don't see a down side to that idea. Might be a cool way to add gain to a typical fender schematic with out drilling the chassis for a tube socket and worrying about heater supply issues.
As much as I consider myself a "tube nut" I can clearly see the benifits of running an FET in the first stage. Good clean stage with no requiered heater current, also microphonics would be less of a concern.
I don't see a down side to that idea. Might be a cool way to add gain to a typical fender schematic with out drilling the chassis for a tube socket and worrying about heater supply issues.
Sorry I wasn't clear, I was referring to the OP seemingly opposite goals.
Picked the FET/tube reference while reading your post, understood you were backing his idea too, that's why I mentioned it.
Personally, I'm very happy with mixing FETs and tubes, if used properly.
Hey!, even using, say, an Op Amp driving a tube, as long as the Op Amp does not clip.
Not prejudiced against using "sand" at all , just recognizing (and respecting) the differences.
What I find strange is adding a clean tube (which is quite "transparent") to the mix, and then simulating distortion digitally.
I think (may be wrong, not really my area of expertise at all) , that if you can simulate a complex, "dynamic", "organic" , you-name-it heavy distortion, which presumably "reacts to pick pressure", is "warm, not buzzy", etc. , all of which I find a difficult goal to achieve, by comparison to add a mild , say, 1% to 5% square law distortion in the algorithm (what usually a "clean" triode stage provides) should be a piece of cake.
And would save on chassis real estate, filament and +V requirements, etc.
Hope this long winded explanation shows my doubts about this "mixed" approach.
Picked the FET/tube reference while reading your post, understood you were backing his idea too, that's why I mentioned it.
Personally, I'm very happy with mixing FETs and tubes, if used properly.
Hey!, even using, say, an Op Amp driving a tube, as long as the Op Amp does not clip.
Not prejudiced against using "sand" at all , just recognizing (and respecting) the differences.
What I find strange is adding a clean tube (which is quite "transparent") to the mix, and then simulating distortion digitally.
I think (may be wrong, not really my area of expertise at all) , that if you can simulate a complex, "dynamic", "organic" , you-name-it heavy distortion, which presumably "reacts to pick pressure", is "warm, not buzzy", etc. , all of which I find a difficult goal to achieve, by comparison to add a mild , say, 1% to 5% square law distortion in the algorithm (what usually a "clean" triode stage provides) should be a piece of cake.
And would save on chassis real estate, filament and +V requirements, etc.
Hope this long winded explanation shows my doubts about this "mixed" approach.
Started reading this and came to a post with the same ideas I would use, then realized it was my own post from a year ago. I found that funny.
So back to the circuit. Doesn't have to be a FET. The whole idea is to hit the first tube a little harder to get some distortion in that stage and then continue on with the next stage. It can be a transistor, opamp, does not really matter, just so the guitar does not get loaded down and you get a hotter signal to the tube. If you got real hot humbuckers it may not be nedded, just depends on where you say enough is enough.
Distortion sounds better when it is caused by multiple stages not all the distortion coming from one stage. You can use the first triode and hit the second with a big signal but sonically it is not the best way to go. A pot in the cathode circuit would work but you do not need the fine control.
So back to the circuit. Doesn't have to be a FET. The whole idea is to hit the first tube a little harder to get some distortion in that stage and then continue on with the next stage. It can be a transistor, opamp, does not really matter, just so the guitar does not get loaded down and you get a hotter signal to the tube. If you got real hot humbuckers it may not be nedded, just depends on where you say enough is enough.
Distortion sounds better when it is caused by multiple stages not all the distortion coming from one stage. You can use the first triode and hit the second with a big signal but sonically it is not the best way to go. A pot in the cathode circuit would work but you do not need the fine control.
It has been over a year since I started on the project. Funny how that works. get started on something and never get back to it. I need to get a life so I can work on projects, now that is the life. I do hope I can get to this and some other projects of interest. I cant talk much more and give lip service as I must build and show the results. One day a week is not much time and then the day is already shot because of other events I have no control over. Have to go out of town for family birthday events, or always something. Oh I have been sick with the flu bug for almost a month. Burned up some of my vacation time being sick and sitting in a chair in front of the TV. You know the story, I am sure I am not alone. frustrating to say the least. I might be selfish I dont know, but I sure am looking forward to working on some projects. I might have to take some time off from work and home and head for the hills with my camper, soldering iron and a bag of parts. I will report back with the results or failure. Actually I think I started on this last November before this one. so 14 months ago. makes me wonder how that can be That I couldn't find even enough time in the smallest degree for this. Oh well have to keep pressing forward.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.