I suspect the OP is after extra harmonic distortion and soft limiting, the effects often called "tube sound."
I like to build stuff. In the long run, I will make both, a neutral one and an effects box just to see how they compare. Right now, I am looking for a reasonably neutral one.
Why I want to make a tube one is because I have nothing with tubes, its all ss. Once this is ready, I will move the B1 to the other system.
Now my turn to ask questions
, since its either effects box or a SS like neutral without tube characteristics (read 2nd harmonics), why do you guys use tubes. Why not just use solid state and save a lot of cash, its easier too, no output transformers, no high voltages etc. If a gear is going to be indistinguishable from SS, why use tubes et al.
Why I want to make a tube one is because I have nothing with tubes, its all ss. Once this is ready, I will move the B1 to the other system.
Now my turn to ask questions
, since its either effects box or a SS like neutral without tube characteristics (read 2nd harmonics), why do you guys use tubes. Why not just use solid state and save a lot of cash, its easier too, no output transformers, no high voltages etc. If a gear is going to be indistinguishable from SS, why use tubes et al.
I wouldn't go so far as to say indistinguishable from SS, unless one cranks in large amounts of global negative feedback. The amps I've built so far are reasonably neutral, but are quite different from my ss amps.
I build tube amps because (1) I used to play around with tubes when I was young and I like the nostalgia, (2) I find studying tube theory to be interesting, so it keeps me thinking, (3) I like to do things a little differently.
Also, its not just second harmonic distortion that plays a part, but a broad spectrum of harmonics and their amplitude relationship that determines the characteristics of the amp.
I build tube amps because (1) I used to play around with tubes when I was young and I like the nostalgia, (2) I find studying tube theory to be interesting, so it keeps me thinking, (3) I like to do things a little differently.
Also, its not just second harmonic distortion that plays a part, but a broad spectrum of harmonics and their amplitude relationship that determines the characteristics of the amp.
Last edited:
First of all, it's a pretty rare guitar effects box that uses really good quality parts and is flat 20HZ - 20kHZ. I wouldn't go that way for my Hi-Fi. Tube amps are art objects. People like looking at them. They typically have a distortion spectrum that is arguably more natural, perhaps more like how the ear-brain mechanism distorts. The distortion as measured by conventional bench techniques seems low enough that there shouldn't be any noticeable difference, yet many say otherwise.
If you put a tube in the signal path that is outside any kind of feedback loop, you'll have its more natural distortion spectrum dominating the distortion spectrum of most decent transistor products elsewhere in the signal path. If tube distortion warms up the sound, this should get you most of the way there.
Personally I'd use 6SN7's with current sources in place of the plate resistor in the case of a gain stage, or in place of the cathode resistor in a follower buffer stage. Not only does the current source increase gain (when applicable), linearize the tubes transfer curve (less distortion), possibly reduce noise, but it gives the stage much better power supply rejection factor, which means less hum.
Personally I'd put tone controls in any preamp I built, which is why I'd have voltage gain stages. Those who think tone controls are evil must not have had a decent one. Shelving is wrong. Baxandall is good. A four section Baxandall is excellent.
The early "James" style topologies would go between gain stages. Since the gain stages typically had no negative feedback (which can cause instability (spurious oscillations)), some think they might sound better. Most of the classics from the 1950's used this approach. Starting in about 1960 most companies switched over to the "Baxandall" topology which puts the tone circuit elements in a feedback loop around one or two tubes that are set up to have substantial voltage gain. These gain stages had feedback through the tone control parts, and sometimes separate from that. If you hate using negative feedback, you might want to try the James type.
Duncan makes a decent circuit simulator for the James topology. You vary the elements and it shows the resulting frequency response. It's a freebie. Google should get you to his page. All the classic early guitar amps used a variation of the James. The Baxandall is easier to make be accurate (flat when pots are centered). The James requires some tweaking to get flat when pots are centered.
As a musician, I can't imagine ever NOT having tone controls. It's about what sounds good, not what is imagined to be more accurate based on lab tests or whatever. Nobody listens to music in an anechoic chamber, and recording engineers spew all kinds of variations of tonality. The way any speaker interacts with the listening room acoustics will have a large effect on tonality. It's typically pretty significant. Pretending flat is always best is foolish.
If you put a tube in the signal path that is outside any kind of feedback loop, you'll have its more natural distortion spectrum dominating the distortion spectrum of most decent transistor products elsewhere in the signal path. If tube distortion warms up the sound, this should get you most of the way there.
Personally I'd use 6SN7's with current sources in place of the plate resistor in the case of a gain stage, or in place of the cathode resistor in a follower buffer stage. Not only does the current source increase gain (when applicable), linearize the tubes transfer curve (less distortion), possibly reduce noise, but it gives the stage much better power supply rejection factor, which means less hum.
Personally I'd put tone controls in any preamp I built, which is why I'd have voltage gain stages. Those who think tone controls are evil must not have had a decent one. Shelving is wrong. Baxandall is good. A four section Baxandall is excellent.
The early "James" style topologies would go between gain stages. Since the gain stages typically had no negative feedback (which can cause instability (spurious oscillations)), some think they might sound better. Most of the classics from the 1950's used this approach. Starting in about 1960 most companies switched over to the "Baxandall" topology which puts the tone circuit elements in a feedback loop around one or two tubes that are set up to have substantial voltage gain. These gain stages had feedback through the tone control parts, and sometimes separate from that. If you hate using negative feedback, you might want to try the James type.
Duncan makes a decent circuit simulator for the James topology. You vary the elements and it shows the resulting frequency response. It's a freebie. Google should get you to his page. All the classic early guitar amps used a variation of the James. The Baxandall is easier to make be accurate (flat when pots are centered). The James requires some tweaking to get flat when pots are centered.
As a musician, I can't imagine ever NOT having tone controls. It's about what sounds good, not what is imagined to be more accurate based on lab tests or whatever. Nobody listens to music in an anechoic chamber, and recording engineers spew all kinds of variations of tonality. The way any speaker interacts with the listening room acoustics will have a large effect on tonality. It's typically pretty significant. Pretending flat is always best is foolish.
Last edited:
Hi, I do not use a common negative connection
only local, as a necessity, transistor noise
never be compared with the tube, the lamp itself is a geometric
electrode shape, no two identical frequency and amplitude
response will always be different, and therefore the sound,
Art to bring you a minimum of feedback,
this is the Hi-Fi. I do not believe in simple circuits "the smaller details
the better "is out of ignorance of physics and electrical engineering.
In a good model, all in place, the remaining 50% is the selection
lamps to minimize noise and background (leakage between the filament and cathode)
saw this making corrector RIAA.
only local, as a necessity, transistor noise
never be compared with the tube, the lamp itself is a geometric
electrode shape, no two identical frequency and amplitude
response will always be different, and therefore the sound,
Art to bring you a minimum of feedback,
this is the Hi-Fi. I do not believe in simple circuits "the smaller details
the better "is out of ignorance of physics and electrical engineering.
In a good model, all in place, the remaining 50% is the selection
lamps to minimize noise and background (leakage between the filament and cathode)
saw this making corrector RIAA.
Bravo, you discovered America!For the best "tube sound" buy a guitar pedal
I asked for advice on how to get a subjective evaluation of the tube/valve sound and you gave this answer,...
Check out the Broskie Akido boards - quite popular around here so lots of help is available.
GlassWare Line Stage & Headphone Amplifiers Kits and PCBs
another option: search for the diytube Budgie PCBs that can be used as a phonostage or low gain linestage. Wallwart power supplies.
diytube.com :: View topic - diy*tube budgie PCBs for sale (Updated 10/28/11)
on the more expensive side, I had good results with the Ultrapath transformer-coupled linestage, but that requires a good clean power supply. At least the gain is low.
Electra-Print.com The Ultrapath
One of my favorite - though very high gain - linestages is a triode-connected EF86 dc-coupled to a 6922 white cathode follower. It has enough oomph to drive a pair of (high-z) headphones. However as stated, the gain is fairly high, so it's best used with an amplifier with low sensitivity.
6th Street Bridge: Triode Connected EF86 linestage and headphone amplifier
for a truly "tubey" sound - a Dynaco PAS or any vintage preamplifier with out-of-spec parts will suffice
GlassWare Line Stage & Headphone Amplifiers Kits and PCBs
another option: search for the diytube Budgie PCBs that can be used as a phonostage or low gain linestage. Wallwart power supplies.
diytube.com :: View topic - diy*tube budgie PCBs for sale (Updated 10/28/11)
on the more expensive side, I had good results with the Ultrapath transformer-coupled linestage, but that requires a good clean power supply. At least the gain is low.
Electra-Print.com The Ultrapath
One of my favorite - though very high gain - linestages is a triode-connected EF86 dc-coupled to a 6922 white cathode follower. It has enough oomph to drive a pair of (high-z) headphones. However as stated, the gain is fairly high, so it's best used with an amplifier with low sensitivity.
6th Street Bridge: Triode Connected EF86 linestage and headphone amplifier
for a truly "tubey" sound - a Dynaco PAS or any vintage preamplifier with out-of-spec parts will suffice
I used an EF86 (the JJ version) in the front end of a guitar amp I built, and it has a voltage gain of about 240, which was more than I wanted. The good news is that a resistive attenuator at it's output might cut noise from the tube more than the noise it ads.One of my favorite - though very high gain - linestages is a triode-connected EF86 dc-coupled to a 6922 white cathode follower. It has enough oomph to drive a pair of (high-z) headphones. However as stated, the gain is fairly high, so it's best used with an amplifier with low sensitivity.
6th Street Bridge: Triode Connected EF86 linestage and headphone amplifier
Hi kstagger!Check out the Broskie Akido boards - quite popular around here so lots of help is available.
GlassWare Line Stage & Headphone Amplifiers Kits and PCBs
another option: search for the diytube Budgie PCBs that can be used as a phonostage or low gain linestage. Wallwart power supplies.
diytube.com :: View topic - diy*tube budgie PCBs for sale (Updated 10/28/11)
on the more expensive side, I had good results with the Ultrapath transformer-coupled linestage, but that requires a good clean power supply. At least the gain is low.
Electra-Print.com The Ultrapath
One of my favorite - though very high gain - linestages is a triode-connected EF86 dc-coupled to a 6922 white cathode follower. It has enough oomph to drive a pair of (high-z) headphones. However as stated, the gain is fairly high, so it's best used with an amplifier with low sensitivity.
http://6streetbridge.blogspot.com/2008/11/triode-connected-ef86-linestage-and.html6th Street Bridge: Triode Connected EF86 linestage and headphone amplifier
for a truly "tubey" sound - a Dynaco PAS or any vintage preamplifier with out-of-spec parts will suffice
How would I create the only preamp with tube EF86 http://6streetbridge.blogspot.com/2008/11/triode-connected-ef86-linestage-and.html first stage and the output capacitor of 2.2uF-MKP.
Or to put the output buffer with ECC88 (6922) first stage after the EF86?
What I got gein and that the output impedance Zout for first stage?
thank you!
Last edited:
hence the triode connection - let me find the original schematic:
The E80F preamp
circuit was revisited here for an alternative cathode resistor:
http://www.klausmobile.narod.ru/testerfiles/6j32p.htm
Last edited:
I liked what i read about ACF-2, thinking of using ECC88s for that. Good choice for a tube newbie build? Will be doing a p2p for that. Any better designs for my needs from a sq perspective? dont need gain.
would like to use a tube rectifier for that. Will try to get chokes, if not, caps it will have to be. Any good existing power supply designs for that here. What tubes will be suitable for the rectifier for this build.
would like to use a tube rectifier for that. Will try to get chokes, if not, caps it will have to be. Any good existing power supply designs for that here. What tubes will be suitable for the rectifier for this build.
I tried using the psud2, but got stuck at a few points. Sorry guys, quite a few newbie and idiotic questions here.
1. How do I know my current requirements. The ecc88 has a max plate current of 25ma. Is 25*4=100ma good enough. sounds a little too much to me. What should I aim for.
2. What voltage transformer do I need. the tubes need 200v as per the document. The recommended trafo is 120-0-120v. would that do for me. 120v and 240v trafos are easier to get. other values I might have to custom order. If a custom trafo will give me better results, I am all for it.
3. What tube to use. I can get 6x4, 6x5gt, ez80, ez81 (maybe), py88. I can also get gz34, 5u4g, 5y3 but these might be more expensive. Do I need two fw rectifier tubes for a ct trafo.
4. If I want to use a choke, whats a good value to start with.
5. In psud2, what am I looking for. I did create a clcrcrcrc supply for playing around, but which current or voltage fields in the graphs should i look at to determine if my design is any good or not. There are quite a few, one per component. Is V(I1) and I(I1) enugh). What should be my load.
6. For heater, I can get 5,6,7,8,9... volt trafos locally easily. I could also get a 6.3v trafo also from an online vendor. Is ac better or dc better for this app. How many amps do I need for this, is 5A good enough. would 3A do.
7. Whats the significance of trafo resistance in psud2. is that specified to calculate current and hence power of the trafo.
8. The resistance for the caps - is that the esr. Even after i get the caps, i might not know the value for this.
1. How do I know my current requirements. The ecc88 has a max plate current of 25ma. Is 25*4=100ma good enough. sounds a little too much to me. What should I aim for.
2. What voltage transformer do I need. the tubes need 200v as per the document. The recommended trafo is 120-0-120v. would that do for me. 120v and 240v trafos are easier to get. other values I might have to custom order. If a custom trafo will give me better results, I am all for it.
3. What tube to use. I can get 6x4, 6x5gt, ez80, ez81 (maybe), py88. I can also get gz34, 5u4g, 5y3 but these might be more expensive. Do I need two fw rectifier tubes for a ct trafo.
4. If I want to use a choke, whats a good value to start with.
5. In psud2, what am I looking for. I did create a clcrcrcrc supply for playing around, but which current or voltage fields in the graphs should i look at to determine if my design is any good or not. There are quite a few, one per component. Is V(I1) and I(I1) enugh). What should be my load.
6. For heater, I can get 5,6,7,8,9... volt trafos locally easily. I could also get a 6.3v trafo also from an online vendor. Is ac better or dc better for this app. How many amps do I need for this, is 5A good enough. would 3A do.
7. Whats the significance of trafo resistance in psud2. is that specified to calculate current and hence power of the trafo.
8. The resistance for the caps - is that the esr. Even after i get the caps, i might not know the value for this.
Here's the image for the psud2. I am also getting a couple of errors.
A current sink has pulled the voltage below zero for more then 5 mains cycles at time 0.100
The rectifier IRFM of 0.25a has been exceeded with a value of 0.25a at time 0.58
A current sink has pulled the voltage below zero for more then 5 mains cycles at time 0.100
The rectifier IRFM of 0.25a has been exceeded with a value of 0.25a at time 0.58
An externally hosted image should be here but it was not working when we last tested it.
1. Ask the preamp designer what current it requires - in some cases it may be marked on the circuit diagram.doors666 said:
2. 120-0-120 will give you about 160V DC if very lightly loaded, and maybe 120-140V when loaded more heavily. This is a bit low, but some circuits may be modified to cope.
3. Depends on current draw. No point in using a big fat hungry rectifier if a smaller one will do the job just as well. Do you have a separate 5V winding? If not, some rectifiers are off limits anyway.
4. 5H or 10H.
5. PSUD2 is a tool. Like any tool you need to know what you are doing. You should certainly be interested in the voltage at the PSU output, and the RMS current through the transformer.
6. Octal rectifiers typically need 5V at 2A from a separate winding with good insulation, as the heater is connected to the cathode. Preamp valves typically take 0.3A or 0.6A at 6.3V. For a line stage AC heating should be fine if you learn how to do proper heater wiring. DC heating can actually be harder to get right for a newbie.
7. Transformer resistance has two main effects: it heats the transformer, it drops the voltage.
8. Don't worry too much about ESR at this stage. It is a detail.
Thanx a lot DF96, I really appreciate your help.
Can I use 25ma per triode * 4 = 100ma. add some margin - 150ma as the current requirements here. Or is it too much of an overkill.
The design is aikido ACF-2 using ecc88, the doc is here. The current is not listed anywhere. If I go by the table at the end of the doc, dividing B+ by Rp gives me values higher than max plate current, that cant possibly be right.
Can I use 25ma per triode * 4 = 100ma. add some margin - 150ma as the current requirements here. Or is it too much of an overkill.
Yeah, and that probably is with ss rectifier. With tubes and chokes, that probably will drop further. The document though mentions B+ as 100, 150, 200, 250 and 300v (in the table at the bottom of the doc), so I still might be good.
Got it. that means all tubes starting with 5 are out. EZ81 looks like a decent choice, can give about 150ma.
What about the current and resistance. I guess the current limits should be well clear of the target current right, so something like 200ma might do in this case. The dcr should be as low as possible to avoid heating and voltage drop. what would be a good value or upper limit for this.
Ok, will use ac heating. Is there an impact if I use 6V instead of 6.3v.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.