If you are using ordinary guitar speakers they will definitely color the tone (frequency wise), probably more than any power amp would ever do. Look at the frequency response of guitar speakers - they are anything but flat! The speaker system is an important part of the amplifier setup and can change the tone and your (stage) volume/headroom dramatically! Just think of it: How do a set of guitar 2 x 15" speakers with high SPL rating, roll-off frequency at 5kHz and free air resonance at 80 Hz compare with a single home stereo grade 6" speaker with a roll-off freq at 20 kHz, resonance at 120Hz and low SPL rating.
Then again, what really is "tube sound"? Is it the sound of the speaker system, the sound of a different transfer curve, the sound of having a transformer coupled circuit or just a nice soundingf tonestack circuit? Think about it.
I have heard so many guitar tones that i lost count. Many people are tricked to believe some amps have tubes just when you plug a compressor into the signal chain. Some people even believe my 12W SS amp is a tube amp when i have the right patch in my digital multieffect. You want overdrive tone and think you get it with just putting a tube to the signal chain? No way! You have to at least clip, pre EQ and post EQ - probably in few stages. Im quite sure you also have to compress before each stage too - at least that's what some hi gain amps seem to do. Blues or 60's tone? - Limiter and mildly overdriven discrete stage (that introduces a bigger amount of 2nd harmonics) - solid state or valve - shouldn't make a big difference. Just look inside some distortion pedals and think a moment what's inside of them...
All and all, you have to know what you need and make a circuit that provides it. Valve amps can be very linear and clip harshly if they are designed to do so - most of them are not. Why do you think opamp's clip harshly; because they are designed to be as linear as possible until the very bitter end. If you feed a very linear circuit with a full bandwidth signal it will not sound very nice - it doesn't make a difference whether you have a valve or a transistor as the active device. If you overdrive a single common emitter stage with a limited, hi-pass filtered signal and after that low-pass filter it you will get a vastly better tone. Valve amplifiers tend to do some of this automatically, however the resulting tone is not very "clean" and thats why people prefer separate distortion/clean channels. By the way, that common emitter circuit probably would have a dominant 2nd harmonic because it clips asymmetrically. Yes, transistor circuits clip asymmetrically and have a dominant 2nd harmonic as well as tube circuits - if you give a chance for that.
You really have to know what you need before yo know what questions to ask. Think about it. I hope i provided some help, not just confusion.
Teemu K
Then again, what really is "tube sound"? Is it the sound of the speaker system, the sound of a different transfer curve, the sound of having a transformer coupled circuit or just a nice soundingf tonestack circuit? Think about it.
I have heard so many guitar tones that i lost count. Many people are tricked to believe some amps have tubes just when you plug a compressor into the signal chain. Some people even believe my 12W SS amp is a tube amp when i have the right patch in my digital multieffect. You want overdrive tone and think you get it with just putting a tube to the signal chain? No way! You have to at least clip, pre EQ and post EQ - probably in few stages. Im quite sure you also have to compress before each stage too - at least that's what some hi gain amps seem to do. Blues or 60's tone? - Limiter and mildly overdriven discrete stage (that introduces a bigger amount of 2nd harmonics) - solid state or valve - shouldn't make a big difference. Just look inside some distortion pedals and think a moment what's inside of them...
All and all, you have to know what you need and make a circuit that provides it. Valve amps can be very linear and clip harshly if they are designed to do so - most of them are not. Why do you think opamp's clip harshly; because they are designed to be as linear as possible until the very bitter end. If you feed a very linear circuit with a full bandwidth signal it will not sound very nice - it doesn't make a difference whether you have a valve or a transistor as the active device. If you overdrive a single common emitter stage with a limited, hi-pass filtered signal and after that low-pass filter it you will get a vastly better tone. Valve amplifiers tend to do some of this automatically, however the resulting tone is not very "clean" and thats why people prefer separate distortion/clean channels. By the way, that common emitter circuit probably would have a dominant 2nd harmonic because it clips asymmetrically. Yes, transistor circuits clip asymmetrically and have a dominant 2nd harmonic as well as tube circuits - if you give a chance for that.
You really have to know what you need before yo know what questions to ask. Think about it. I hope i provided some help, not just confusion.
Teemu K
Shoog said:
This sounds like an optimal solution. Using one or two tubes and a cheap transformer to make an okay-sounding preamp is ideal; I'm still kind of fuzzy on how tubes work, but I'd at least like to try making something with them.
Also, if I can get an appropriate tube, I think I'll try a stab at the single-tube solution mentioned above. With a spare guitar amp speaker (something cheap, probbably) it might make an interesting, and extremely simple, first amp.
Buy an Epiphone Valve Junior and hook up one of these:
http://www.scopeboy.com/ssim.html
Or really - just go try some of the tube pedals out there. They are popular for good reason.
http://www.scopeboy.com/ssim.html
Or really - just go try some of the tube pedals out there. They are popular for good reason.
The last post by Raintalk looks as if it gives a bit of useful information about optimising this arrangement.
I suggests that you go ahead with a simple single valve preamp stage, a step down interstage and finally a simple LM3875 implementation. This will be cheap and very simple to realise and it will surprise you with the quality of its sound.
You have nothing to loose and you might just create something that others will want to follow.
If you haven't built a chip amp before, try to get hold of a couple of chips in case of accidents. You may have serious DC offset issues if you don't include a simple time delay circuit, this will allow the valve to warm up before it starts driving the chip amp- which will fry your speaker !!!
Shoog
I suggests that you go ahead with a simple single valve preamp stage, a step down interstage and finally a simple LM3875 implementation. This will be cheap and very simple to realise and it will surprise you with the quality of its sound.
You have nothing to loose and you might just create something that others will want to follow.
If you haven't built a chip amp before, try to get hold of a couple of chips in case of accidents. You may have serious DC offset issues if you don't include a simple time delay circuit, this will allow the valve to warm up before it starts driving the chip amp- which will fry your speaker !!!
Shoog
Part deux
Look at the first chart at: http://milbert.com/articles/TvsT/tvtiega.html
Note the frequency response of the tube amp.
This is due to several things, but one factor is the reactive load of the speaker. You can't just hook up a resistor in place of a speaker and expect it to emulate a speaker. You could hook up a resistor and use EQ to mimic the same curve. Or you could use a more reactive dummy load:
http://www.duncanamps.com/technical/dummyload.html
Personally - I'd just use EQ.
Look at the first chart at: http://milbert.com/articles/TvsT/tvtiega.html
Note the frequency response of the tube amp.
This is due to several things, but one factor is the reactive load of the speaker. You can't just hook up a resistor in place of a speaker and expect it to emulate a speaker. You could hook up a resistor and use EQ to mimic the same curve. Or you could use a more reactive dummy load:
http://www.duncanamps.com/technical/dummyload.html
Personally - I'd just use EQ.
teemuk said:
I think this thread is still lacking the definition of the term "tube sound" that the thread starter has been looking for. Yeah, surely the transformer coupled reactive load will make some "EQing" to the signal as well as the tube will have some non-linearities but the question is are they enough. I still have no clue whether the thread starter is after for a "tube" clean tone, "tube" overdriven tone, or perhaps "tube" hi-gain tone. And on top of that, what makes them "tube" for the person who started this thread?
What i'm saying is that anyone will definitely get a "tube sound" just by implementing a tube as the active device... pretty logical, right; it has to be a tube sound since it's coming from a tube. But is it the tone that the thread starter is really after?
You can implement transformers, tubes - all kinds of stuff - and make a circuit that distorts like hell - or stays clean. When you say "tube sound" what do you mean spasticteapot?
Teemuk
teemuk said:
Look, stop pestering the guy. You showed off how much you know about the subject with your lengthy post. Stop the contest and try to offer help to someone at a more basic level.
I think this thread has gotten too philosophical for what it started off as, a beginner (relatively) asking if this was a good path to pursue. Yes, you can get good results with just a tube preamp stage and a chip power stage. There are commercial rackmount tube distortion boxes that have no iron and no output tubes and no Celestion in them AFAIK and can produce a wide range of sounds, as well they should with their 4 figure price tag. Unfortunately, I haven't come across a good published design for one of these yet. The other suggestions in the thread are a best bet.
I quite like Teemuk's answer and question.
I could think that teapot, not being a guitar player, could build something with a tube, take it to his guitar player friends and have them plug in and comment: "Where's the tube sound?" or maybe "That's over the top" or unlikely: "That's perfect"
Maybe the philosophy would trim down if the question were clarified with "Sound like X... Brand amp" Or "Sound like on this cut of music"
?
I could think that teapot, not being a guitar player, could build something with a tube, take it to his guitar player friends and have them plug in and comment: "Where's the tube sound?" or maybe "That's over the top" or unlikely: "That's perfect"
Maybe the philosophy would trim down if the question were clarified with "Sound like X... Brand amp" Or "Sound like on this cut of music"
?
raintalk said:
To be quite honest, I'm a fan of "the wierder, the better". Old tube-based Fender amps are a nice target, though.
I like the idea of the paralell configuration with the transformer; I just don't understand what exactly it does. Could someone explain it in plain english? I'm guessing that it converts voltage to current, dropping the effective gain but letting it actually drive something at more than a few mA, but this is just a wild guess.
I'm not opposed to using more tubes; I'm just opposed to the really, really high price. I might be able to get an old tube amp that's half-blown and rip out the preamp, if I'm lucky.
Also, why not simply make the overdriven tube the same tube that's used as a buffer for the gainclone?
Also, I hate to repeat..but could the above class A amp (which looks like a nice starting point) work with a smaller transistor and if it had the input fed through a gainclone?
With regard to the interstage transformer, your analysis is just about spot on. The transformer undistorted would introduce its own tone to the sound. If overdriven it would distort by saturating, this will involve clipping the waveform which would be a form of distorted compression. My understanding is that some of the older guitar amps used underspecified output transformers to introduce just this effect. Again a little bit of research will tell you which of the older designs used this.
You need to find a simple guitar driver circuit which has a good reputation, and then work out how to make it work with a transformer. Shouldn't be to difficult really.
It is a good point that you need to decide what you want the end result to sound like before you can design that sound.
Shoog
You need to find a simple guitar driver circuit which has a good reputation, and then work out how to make it work with a transformer. Shouldn't be to difficult really.
It is a good point that you need to decide what you want the end result to sound like before you can design that sound.
Shoog
Spasticteapot said:
You nailed it quite right. Look at the fourth paragraph on this page:
http://www.circuit-innovations.co.uk/transformers.html
Also, the transformer will "transform" resistance, capacitance and inductance according to the turns ratio. The small impedance on the secondary side will be seen as a higher impedance on the primary side - high enough to load the tubes properly.
See this page by Randall Aiken, it's a good source of information:
http://www.aikenamps.com/OutputTransformers.html
Now, if you connect a speaker to the secondary side you will not have a constant voltage gain throughout the whole bandwidth - as you would have if you replace the speaker load with a plain resistor. This is one of the things considered being part of "power amplifier distortion" characteristics. If you just transformer couple a tube stage to the chipamp's constant internal resistance you will not get the frequency response of a tube amp driving a speaker load - which is anything but flat.
The thing is, if you want a tube amp sound you have to copy a tube amp. In simplest way this would, for example, mean a common cathode stage driving a load that mimicks the impedance curve of a speaker. If you use a low power tube you can also use very cheap transformers and the lack of enough drive current will distort the signal. (I guess the distorted tone is what you are after - otherwise what's the point of using tubes). Then just feed the (probably attenuated) output voltage for the chipamp power amplifier stage. This will give you a tube amp tone to some extend. If you want more characteristics of your favourite amp circuit you have to mimick it more. I'd suggest copying at least the tonestack.
The amount of tubes will unlikely cost that much extra when compared to "iron": transformers, heatsinks etc. I shouldn't worry too much about it. In most cases one tube actually means two, so using two tubes will give you four gain stages - enough to build a whole preamp and poweramp with lot's of fancy stuff.
If you worry about the price, using a "parafeed" topology (capatively coupled transformer as a load) might be the most cost-saving choice. You can use a smaller and cheaper OT this way. Also, you might want to consider using a lower plate voltage. If you stick with 50 - 60 V secondary voltage you do not need a special and pricey transformer - although you need another one for filaments. This will of course mean lower headroom and more distortion but you can omit using high voltage capacitors which are quite expensive and tough to find.
I had a thought. It would be nice to have a means to change the amount that the transformer was overdriven. One way of doing this is to turn up the volume knob, but this needs for the transformer to be precicely specified to achieve this, and also it need another volume control before the output stage (preferably tied to the first one) so you don't deafen yourself. Complex.
A simpler approach is to use one of the characturistics of parafeed to your advantage. The parafeed cap is there to stop DC from getting into the un-airgapped transformer and saturating it. But we want saturation, so what if we put a pot(and a series resistor) across the parafeed cap to allow a small controlled amount of DC into the transformer so introducing a controlled amount of saturation. Call this our overdrive knob. It would want to be a wire wound pot or a stepped attenuator, otherwise it will soon have bad wiper noise.
On the subject of a class A output stage, I love class A. However they may look simple on paper, but to even achieve 8 watts of output requires a big hot "expensive" piece of kit.
Shoog
A simpler approach is to use one of the characturistics of parafeed to your advantage. The parafeed cap is there to stop DC from getting into the un-airgapped transformer and saturating it. But we want saturation, so what if we put a pot(and a series resistor) across the parafeed cap to allow a small controlled amount of DC into the transformer so introducing a controlled amount of saturation. Call this our overdrive knob. It would want to be a wire wound pot or a stepped attenuator, otherwise it will soon have bad wiper noise.
On the subject of a class A output stage, I love class A. However they may look simple on paper, but to even achieve 8 watts of output requires a big hot "expensive" piece of kit.
Shoog
Take a look at the Paia Stack-in-the-box schematic:
http://paia.com/tubestuf.htm
It does the low plate voltage trick.
Read the link at the bottom of the page on "tube" sound.
That stack-in-the-box doesn't have an xformer, but it has a speaker simulator. It looks like some kind of simulated inductor circuit.
You could probably add in a real xformer - with a switch to flip it in and out.
I don't know about pushing DC through the primary to force saturation. A transformer has two sides, and the load side affects how the primary side reacts, and the effects of saturation. So it seems it would be better to vary the speaker simulator on the output side.
PS - I don't know of too many preamps that aren't class-a
http://paia.com/tubestuf.htm
It does the low plate voltage trick.
Read the link at the bottom of the page on "tube" sound.
That stack-in-the-box doesn't have an xformer, but it has a speaker simulator. It looks like some kind of simulated inductor circuit.
You could probably add in a real xformer - with a switch to flip it in and out.
I don't know about pushing DC through the primary to force saturation. A transformer has two sides, and the load side affects how the primary side reacts, and the effects of saturation. So it seems it would be better to vary the speaker simulator on the output side.
PS - I don't know of too many preamps that aren't class-a
raintalk said:Take a look at the Paia Stack-in-the-box schematic:
http://paia.com/tubestuf.htm
It does the low plate voltage trick.
Read the link at the bottom of the page on "tube" sound.
That stack-in-the-box doesn't have an xformer, but it has a speaker simulator. It looks like some kind of simulated inductor circuit.
You could probably add in a real xformer - with a switch to flip it in and out.
I don't know about pushing DC through the primary to force saturation. A transformer has two sides, and the load side affects how the primary side reacts, and the effects of saturation. So it seems it would be better to vary the speaker simulator on the output side.
PS - I don't know of too many preamps that aren't class-a
Interesting design. I don't, however, need the simulator; I'm using a real guitar speaker driver.
Any reason why the line-out preamp I originally listed would not work?
Spasticteapot said:
Sure it's got tubes - so you'll get a "tube sound".
I don't know that it gives Fender Bassman "tube sound". You can find Fender Bassman preamp schematics on the web, but what's in controversy is can just a preamp give "tube sound". But give it a try, If it's not getting the sound you want you can tweak it with some of the techniques given by others here.
On the simulator, I think what's being said is that by driving the output of a tube into a circuit that simulates a speaker you're more likely to get "tube sound" because the speaker interaction is part of the "tube sound". In your case you'll be driving the speaker from a power amp buffer so it can't interact with the tube circuit. You may want to add simulation right after the tubes and before the power amp.
It's also being said that you need to drive a xformer to get xformer part of "tube sound". I'm still trying to understand this myself.
"t's also being said that you need to drive a xformer to get xformer part of "tube sound". I'm still trying to understand this myself. "
I have never said that it can't be done any other way, what I have suggested is that using an interstage has certain advantages (ie impedance matching) and will produce a type of distortion which is particular to a transformer, and would be difficult to simulate otherwise. The impedance matching is worth it in itself as far as I am concerned.
I personally don't think a simple tube based preamp will come close to the sound our friend is looking for.
Of course as always I could be totally wrong!
Shoog
I have never said that it can't be done any other way, what I have suggested is that using an interstage has certain advantages (ie impedance matching) and will produce a type of distortion which is particular to a transformer, and would be difficult to simulate otherwise. The impedance matching is worth it in itself as far as I am concerned.
I personally don't think a simple tube based preamp will come close to the sound our friend is looking for.
Of course as always I could be totally wrong!
Shoog
Just some more useful thoughts about the use of a parafeed arrangement. The parafeed cap and the transformer are a resonant circuit. This resonance is usually tuned to subsonic levels, but we can tune this to any frequency we choose. By using a cap in the range of 0.1 - 2uf (to be determined by experiementation) we should be able to get it to resonate in the bass range - mimicking the impedence rise of a guitar speaker. It might even be possible to make this variable by a simple cap resistor network in parallel with the main parafeed cap.
Also by choosing a cheap EI transformer (think split bobbin) we can introduce high interwinding capacitance, which should mimic the rising impedence of a guitar speaker at higher frequencies.
I am personally so excited about the possabilities that I am tempted to take up the challenge to build a friend a "cheap" valve sounding practice amp. Should be possible for less than €150.00 and a little imaginative parts salvaging. I already have a stack of LM3785's with no use.
Shoog
Also by choosing a cheap EI transformer (think split bobbin) we can introduce high interwinding capacitance, which should mimic the rising impedence of a guitar speaker at higher frequencies.
I am personally so excited about the possabilities that I am tempted to take up the challenge to build a friend a "cheap" valve sounding practice amp. Should be possible for less than €150.00 and a little imaginative parts salvaging. I already have a stack of LM3785's with no use.
Shoog
Shoog, your idea sounds workable. I personally have built hybrids, and overdriving the tube preamp doesnt work. I am thinking of trying your interstage idea... My objective is to be cheap (thats why hybrid in first place), so it would be great if any off-the-shelf power transformers can be used. Any suggestions?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.