Hi All
A while ago I put together an op-amp based preamp for my acoustic guitar pickup with some fantastic help from several forum members here. This works great and I am very pleased with the sound. I got to thinking about a simple tube preamp stage and wondering if this would work with an acoustic transducer. The only tube I have is a brand new, never been kissed ECC83 and so got searching for a single tube preamp . I found this preamp stage...
DIY 12AX7 Tube preamp
I know nothing of tube techniques and so wondered if you guys and girls could help me out. Is this circuit suitable for my needs? which are...
1. An input impedance of 1 Megohm (crucial!)
2. About 18dB gain maximum
3. An output suitable to drive a 20K to 40K amplifier input
4. Capable of handling around 250 to 300mV rms input max with very
low distortion and good fidelity
5. Using one ECC83
Can you good people tell me if the circuit is any good or can be modified to suit my needs. If not, do you know of any such circuit. I will design a build a custom linear power supply for the voltages and currents required for the unit but I would need a circuit that would be up and running with the minimum “fiddling” as my resources and time to devote is always limited!
One more thing, the components in the board photos in the above link do not seem to agree with the schematic. Is there anything I am missing here?
I am hoping you guys and girls can come to my aid!
Cheers
Ray
A while ago I put together an op-amp based preamp for my acoustic guitar pickup with some fantastic help from several forum members here. This works great and I am very pleased with the sound. I got to thinking about a simple tube preamp stage and wondering if this would work with an acoustic transducer. The only tube I have is a brand new, never been kissed ECC83 and so got searching for a single tube preamp . I found this preamp stage...
DIY 12AX7 Tube preamp
I know nothing of tube techniques and so wondered if you guys and girls could help me out. Is this circuit suitable for my needs? which are...
1. An input impedance of 1 Megohm (crucial!)
2. About 18dB gain maximum
3. An output suitable to drive a 20K to 40K amplifier input
4. Capable of handling around 250 to 300mV rms input max with very
low distortion and good fidelity
5. Using one ECC83
Can you good people tell me if the circuit is any good or can be modified to suit my needs. If not, do you know of any such circuit. I will design a build a custom linear power supply for the voltages and currents required for the unit but I would need a circuit that would be up and running with the minimum “fiddling” as my resources and time to devote is always limited!
One more thing, the components in the board photos in the above link do not seem to agree with the schematic. Is there anything I am missing here?
I am hoping you guys and girls can come to my aid!
Cheers
Ray
Hi Ray,
it looks like that circuit fulfills all your requirements, except that it appears to have a gain of 51, which is about 34 dB while you only need 18 dB.
Of course, the ECC83 being a high-mu tube, most schematics you encounter will give more than that 18 dB gain unless some sort of feedback is incorporated.
Kenneth
it looks like that circuit fulfills all your requirements, except that it appears to have a gain of 51, which is about 34 dB while you only need 18 dB.
Of course, the ECC83 being a high-mu tube, most schematics you encounter will give more than that 18 dB gain unless some sort of feedback is incorporated.
Kenneth
Hi, from simplest to most involved, there's a number of things you could do:
- ignore the problem, maybe your main amp turns out to have enough headroom?
- put a 10 Kohms trimmer potentiometer across the output so you can trim the output level
- incorporate some form of NFB into the circuitry
Probably best to try it in that order, I think the last option would be overkill.
Best,
Kenneth
- ignore the problem, maybe your main amp turns out to have enough headroom?
- put a 10 Kohms trimmer potentiometer across the output so you can trim the output level
- incorporate some form of NFB into the circuitry
Probably best to try it in that order, I think the last option would be overkill.
Best,
Kenneth
I'd go for the 6922 version of that pré-amp:
DIY 6922 / E88CC Tube preamp
Maybe include a tonestack between gain and follower stage...
(this would lower gain to the amount you want)
Feedback will not help "the cause" since you will alter the balance between even and odd harmonics, imho, not what an acoustic guitar needs...
DIY 6922 / E88CC Tube preamp
Maybe include a tonestack between gain and follower stage...
(this would lower gain to the amount you want)
Feedback will not help "the cause" since you will alter the balance between even and odd harmonics, imho, not what an acoustic guitar needs...
If you put together a valve stage with "very low distortion and good fidelity" then it should sound exactly like your op-amp circuit. People often use valves, especially with guitars, because they want to avoid very low distortion (although they may not always admit to this!). I think you need to be clearer about what you are aiming at: fidelity or "tube sound". Either can be achieved using valves, but the design would be different.
Thanks for all your replies!
Ken - seems like putting a volume pot on the output would be a solution, but wouldn't this raise the output impedance a little high for a 20K amp input. Also this output impedance would vary with the setting of the control?
Oneyed – Looks like feedback would not be good. Can you fill me in on what a Tone
Stack is and what it involves as I am a bit of a newb to all this (blush!)? The E8CC tube looks like it would be nicer but a little pricey (prices vary wildly and can go through the roof!!). I have an ECC83 to hand
DF96 – Good point! I think maybe I would like to try the tube sound on the acoustic and just settle for the best fidelity that a simple arrangement can give. Hopefully a little warmth (circuit distortion/compression etc) may just make for a good acoustic sound?
Thanks for your help
Ray
Ken - seems like putting a volume pot on the output would be a solution, but wouldn't this raise the output impedance a little high for a 20K amp input. Also this output impedance would vary with the setting of the control?
Oneyed – Looks like feedback would not be good. Can you fill me in on what a Tone
Stack is and what it involves as I am a bit of a newb to all this (blush!)? The E8CC tube looks like it would be nicer but a little pricey (prices vary wildly and can go through the roof!!). I have an ECC83 to hand
DF96 – Good point! I think maybe I would like to try the tube sound on the acoustic and just settle for the best fidelity that a simple arrangement can give. Hopefully a little warmth (circuit distortion/compression etc) may just make for a good acoustic sound?
Thanks for your help
Ray
Since the output of the circuit is around 600 ohms, the pot could be 5K, too.
A little 2nd order distortion (without overdoing it) is definitely pleasantly sounding for acoustic instrument amplification. The distortion can be raised a little by bypassing R9 with a large capacitor, but that will also raise the gain of the circuit, which is exactly the opposite of what you need
Kenneth
A little 2nd order distortion (without overdoing it) is definitely pleasantly sounding for acoustic instrument amplification. The distortion can be raised a little by bypassing R9 with a large capacitor, but that will also raise the gain of the circuit, which is exactly the opposite of what you need
Kenneth
The ECC83 is not a suitable valve for your requirements. It is fairly linear, when used properly, but has high gain. The gain can be reduced by adding negative feedback, but then you are approaching fidelity rather than a little valve distortion.
You need to start with a much lower gain: ECC88 or ECC82. As a first rough stab, take the ECC83 circuit you started with. Leave the 1M resistors as they are, but divide all the others by a factor of 3 or 4. Use an ECC82 - NOS is quite cheap. Give it a try! If it needs more gain then bypass R9.
You need to start with a much lower gain: ECC88 or ECC82. As a first rough stab, take the ECC83 circuit you started with. Leave the 1M resistors as they are, but divide all the others by a factor of 3 or 4. Use an ECC82 - NOS is quite cheap. Give it a try! If it needs more gain then bypass R9.
E88CC JJ Red Label (aka 6922) JJ (former TESLA)
--> not too expensive, I'd say
And yes, ECC82 is also a good choice...
I just like the sound of the ECC88 better, depending how you
play the acoustic guitar, it can generate peaks that the ECC88 will take better than ECC82 or ECC83 (in the last case, presented by generating a more "electric" sound.)
TSC
--> tone stack = small eq
--> not too expensive, I'd say
And yes, ECC82 is also a good choice...
I just like the sound of the ECC88 better, depending how you
play the acoustic guitar, it can generate peaks that the ECC88 will take better than ECC82 or ECC83 (in the last case, presented by generating a more "electric" sound.)
TSC
--> tone stack = small eq
Last edited:
OK, OneEyed and DF96 have got me sold on the E88CC/6922 version, and by the description given of the sound in the project link, it sounds as though it would be about ideal as I do not want an “electric guitar tone”. The 250V HT would be a trifle easier to handle also. Thanks guys!
However the gain is still a little high. I decided to play around in LTspice to see if I could reduce the gain. I thought increasing the cathode resistor value on the first triode stage might do the job and this seems to be the case. I settled on 2.2K and here is the LTSpice schematic...
http://myweb.tiscali.co.uk/raykel/Spice_Schem_6922.JPG
I did some checks and LTSpice (or at least my poor modelling skills) reckons the following...
1. Gain a little under 18.3dB at 1Khz
2. THD=0.015% at 1Khz
3. Input Z=988K, Output Z=approx 124Ohms
4. Phase inversion (could help with acoustic feedback at higher volumes?)
5. HT current a little over 11.mA at 250V
Do you think this circuit is a goer and do you think that it will give enough “tube” signature to be worthwhile building? Also do you think the input 100n input cap, coupled with the input Z, will give a suitable bottom end for acoustic guitar amplification, as I want to capture the depth of the instrument (C.F Martin J model) without too much sub bass frequencies causing feedback problems etc?
I am a little concerned about how smooth the supply voltages need to be as explained in the project link, but I am sure I can come up with a suitable supply with some trial and error
I have attached the LTSpice files should you be interested in having a play!
Appreciate all you help!
Cheers
Ray
However the gain is still a little high. I decided to play around in LTspice to see if I could reduce the gain. I thought increasing the cathode resistor value on the first triode stage might do the job and this seems to be the case. I settled on 2.2K and here is the LTSpice schematic...
http://myweb.tiscali.co.uk/raykel/Spice_Schem_6922.JPG
I did some checks and LTSpice (or at least my poor modelling skills) reckons the following...
1. Gain a little under 18.3dB at 1Khz
2. THD=0.015% at 1Khz
3. Input Z=988K, Output Z=approx 124Ohms
4. Phase inversion (could help with acoustic feedback at higher volumes?)
5. HT current a little over 11.mA at 250V
Do you think this circuit is a goer and do you think that it will give enough “tube” signature to be worthwhile building? Also do you think the input 100n input cap, coupled with the input Z, will give a suitable bottom end for acoustic guitar amplification, as I want to capture the depth of the instrument (C.F Martin J model) without too much sub bass frequencies causing feedback problems etc?
I am a little concerned about how smooth the supply voltages need to be as explained in the project link, but I am sure I can come up with a suitable supply with some trial and error
I have attached the LTSpice files should you be interested in having a play!
Appreciate all you help!
Cheers
Ray
As you have Spice, why not play with an ECC82 version too? It will have lower gain, and a different distortion signature. Don't look at the total distortion, but the 2nd and 3rd separately as these will sound different from each other.
In order to get the ECC88 gain low enough you have had to bias it into a non-linear region, although you will get some negative feedback from the unbypassed cathode resistor. My guess is that this will give more 3rd and less 2nd than an '82 circuit - but see what Spice says.
In order to get the ECC88 gain low enough you have had to bias it into a non-linear region, although you will get some negative feedback from the unbypassed cathode resistor. My guess is that this will give more 3rd and less 2nd than an '82 circuit - but see what Spice says.
Sorry, I can't help you there. I have B2 Spice. I'm sure someone can help.
One way to check which is the dominant harmonic order is to see how THD changes with signal level. 2nd means THD is proportional to signal. 3rd means THD goes like signal squared. At low signal levels 2nd will always dominate anyway.
One way to check which is the dominant harmonic order is to see how THD changes with signal level. 2nd means THD is proportional to signal. 3rd means THD goes like signal squared. At low signal levels 2nd will always dominate anyway.
Thanks DF96! I am not sure how to go about getting this info in LTSpice, hopefully someone here will be kind enough to offer advice.
Is the 2.2K cathode resistor correct or will this value cause problems? I do not know of any other way of reducing the gain to the 15 to 18dB that I need.
Cheers
Ray
Is the 2.2K cathode resistor correct or will this value cause problems? I do not know of any other way of reducing the gain to the 15 to 18dB that I need.
Cheers
Ray
2.2k will indeed reduce the gain, although it will greatly increase the distortion since you are biasing colder. Not sure whether this would bother you though- I sort of lost track of what you want from this circuit; linearity or classic "valve tone" (if the latter then there's nothing wrong with an ECC83. If the former, why use valves at all..?)
An alternative method of reducing the gain is to split the anode resistor (R1 in your diagram) into two parts and take the output (c1) from the junction. In other words, treat the anode resistor like it is a pot, and play with the values to get the gain you want without altering the bias of the valve. This would work with any valve of course, whatever its maximum gain, and it does not alter the linearity of the circuit.
Last edited:
Hi Merlinb
Thanks for your reply. Ah I didn’t realise that 2.2K would dramatically increase the distortion. I appreciate your confusion in my overall goal, I basically want to see if a tube preamp would provide a benefit to my acoustic guitar pickup. I need enough fidelity and clarity to bring out all the nuances of the acoustic tone with both strumming and finger-style playing, but without going full into the “Marhall” front end electric sound. I was hoping that maybe a good tube preamp may add say a little warmth without adding too much distortion. The gain I need is not huge as the pickup is quite high output.
I can appreciate that if the tube preamp was extremely HiFi, then I would not achieve any benefit over my Opamp based preamp (TL071). So this is why I am asking the tube gurus for advice on what circuit they think would suit acoustic guitar amplification best. Due to family pressures etc, I unfortunately do not have the time and resources to do too much “suck it and see” research (unlike some of you lucky pups!!). Ideally I would like to have a fairly good idea that a circuit will be suitable before I go ahead and build.
I hope this does not cause even more confusion and thanks again for your reply
Cheers
Ray
Thanks for your reply. Ah I didn’t realise that 2.2K would dramatically increase the distortion. I appreciate your confusion in my overall goal, I basically want to see if a tube preamp would provide a benefit to my acoustic guitar pickup. I need enough fidelity and clarity to bring out all the nuances of the acoustic tone with both strumming and finger-style playing, but without going full into the “Marhall” front end electric sound. I was hoping that maybe a good tube preamp may add say a little warmth without adding too much distortion. The gain I need is not huge as the pickup is quite high output.
I can appreciate that if the tube preamp was extremely HiFi, then I would not achieve any benefit over my Opamp based preamp (TL071). So this is why I am asking the tube gurus for advice on what circuit they think would suit acoustic guitar amplification best. Due to family pressures etc, I unfortunately do not have the time and resources to do too much “suck it and see” research (unlike some of you lucky pups!!). Ideally I would like to have a fairly good idea that a circuit will be suitable before I go ahead and build.
I hope this does not cause even more confusion and thanks again for your reply
Cheers
Ray
Hi DF96
Thanks bud. I need 1.0MEG imput impedance for definite as the pickup specification demands this for full frequency response. I suppose an output pot would not be too bad if say a 5K component could be used, as I need the preamp to match a 20K input impedance worst case. I will play around with Merlinb’s split anode resistance and see what happens in LTspice, as this seems like it could be a possible solution
Cheers
Ray
Thanks bud. I need 1.0MEG imput impedance for definite as the pickup specification demands this for full frequency response. I suppose an output pot would not be too bad if say a 5K component could be used, as I need the preamp to match a 20K input impedance worst case. I will play around with Merlinb’s split anode resistance and see what happens in LTspice, as this seems like it could be a possible solution
Cheers
Ray
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.