Maker Pro
Maker Pro

Valve (tube) amplifier schematic help

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Per specs impedance is at 36 ohms.

Not surprising. Many earphones today are actually small speakers. Waaaaaaaaaaaaaaay too low Z for your preamp. I'll post an example of what impedance means to you and why it's important..

Chris

On second thought I don't want to do this unless you understand Ohms Law. Do you?
 
Last edited:

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
I understand the fundamentals of Ohm's Law.

A lot of the various manipulations of it I have no experience with, but give me adequate time and it would make enough sense.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
I'll post it then.

Chris
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
Take a look at what happens to the output signal when the load (RL) is 100KΩ (FigA) and when it's reduced to 36Ω (FigB)

Chris
 

Attachments

  • OutputZ.JPG
    OutputZ.JPG
    73.3 KB · Views: 110

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
The signal voltage is reduced. It went from +/- ~5mV to +/- 3.5 mV.

Is that the correct thing to notice?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
The signal voltage is reduced. It went from +/- ~5mV to +/- 3.5 mV. Is that the correct thing to notice?
You got it. Actually it was +/- ~5V (volts, not mV), so the voltage has dropped by a factor of more than 1000, or 60 dB!

Nice work with that simulation Chris! That shows the effect clearly. On the bottom right trace, which shows the voltage across a 36 ohm transducer, the amplitude has reduced from 10V p-p down to about 7 mV p-p which will be barely audible.

So does that also explain the lack of quality?
Will chaining through to a larger amp unit remedy this?
"Lack of quality" is too vague. It will sound very faint and "thin" because there are only the high frequencies present.

The important factor is the impedance that your preamp is driving into. That preamp is designed to drive a load impedance of 100 kilohms or higher. As you've discovered, 36 ohms is far too low.

I'll try to explain this a bit better in another post.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
Oh okay. It was a thousand-fold drop!

That explains a lot!

The graphs were all in phase and similar, and so it was hard to read. Chris, what program do you use to make the circuit schematics?

Also, in the next hour or so I will scan up another schematic for the new tube amp I am building to have you two check it. This one will (hopefully) be done right.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
circuit-1.jpg


Fingers crossed that I'm on the right track with this one. It's taken me about 10 sheets of graph paper to get this correct.

3x 12A series valves, one for left channel, one for right, one for combined stereo amplification. The two are then split to two IRF510 MOSFETs for power amplification (one being L+S channel, the other R+S channel). The two MOSFETs are chained to the 12VDC plate supply. I decided on the IRF510s because:

1) They're a common component - I can buy them at Radioshack for about two bucks a pop.
2) I assembled and drafted this circuit from a session of running rampant and jacking everyone's designs throughout the internet. From what I can gather, other headphone amps use the IRF510 and similar FETs for power amplification stages.

Now, I'm not an idiot, I don't know any better with this stuff, so while I'm still in my ignorance about solid-state devices, I might as well steal parts of designs from people who know what they're doing ;)

Let me know how everything checks out! The scanner butchered the detail, so if you need anything explained let me know.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
There's no reason to abandon the original design. It just needs to be modified for use with a low impedance load. Speaking of which.... We discussed the that the 12AU7 amp was intended to drive a high Z input guitar amp. This discussion was quite extensive early in the thread. You never interjected to tell us you intended otherwise. :eek:

This can't be repeated too many times. ... I really do think that you should be spending much of your time in our "Tutorials" section.... At the beginning.

Chris
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
There's no reason to abandon the original design. It just needs to be modified for use with a low impedance load. Speaking of which.... We discussed the that the 12AU7 amp was intended to drive a high Z input guitar amp. This discussion was quite extensive early in the thread. You never interjected to tell us you intended otherwise. :eek:

This can't be repeated too many times. ... I really do think that you should be spending much of your time in our "Tutorials" section.... At the beginning.

Chris

Different design, different purpose. Hence the solid state section for secondary amplification.

I'm not abandoning the original design, it has just caused me a month and a half's worth of headache. The issue with the current build the way it is is that the entire enclosure will have to be stripped down and rewired, new supplies drafted and soldered, new circuits soldered with extra emphasis on cutting down lead length to cut down on the ability for demodulation by the circuit. The entire enclosure was fabricated wrong, combine three different pieces off by 2-3mm in places and undercut by 1mm in others and you get something that doesn't fit together right even in best case scenario. There are so many small things that I feel I messed up on on this project that simply redoing all them would probably incur just as much time and expense as it took to get here in the first place. 100% total revamp.

Don't get me wrong, I don't intend to simply bury my mistakes in a new build and never look at the old build again. I just want to start a new build for a different purpose, and the new design elements will keep me from being driven over the edge when a current issue cannot be completed by design flaws I made a month ago. I may not know much about the specifics of electronics, but I know to learn from my mistakes, and I know many things aren't done for a reason. Learning what those reasons are...that will take time.

I have two weeks and one weekend before I go off to college. If I were to tackle this new project it only leaves me time to order the PCB blanks, etch and drill them anyways. I would rather spend this last bit of time on a 'clean slate' then try to rebuild the one that I failed.

Tutorial section...I've lived there for the past two weeks. At least two tabs have been open on my computer every day, and it hasn't been rebooted all summer. I have tried to sap every little bit of knowledge off of every circuit I've looked at in the past few weeks. Every one of my friends is already off at college so every day I've had nothing else to do but post here, look up circuits, lurk other forums, and scribble nonsensical circuits on a whiteboard and then trying to figure out if they would or wouldn't work. Combined with the occasional trip to 7/11 for a change of scenery, or Radioshack to pick up components.

Just let me know what I need to change or show me a revision on that circuit to allow itself for low-impedance use and to make it work right, and I will go on my merry way, let this thread die and try and figure out my own problems if they arise.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
On a slightly more cheery hand,

DSC_0888.jpg


There is hope for me. I have figured a way to light LEDs off 120VAC mains without causing anything to explode or the circuit genie to come possess me.

Thing I've realized about myself is getting up in the morning and jumping on the computer often leads to incoherent thoughts and things not coming through as they should. If my tone in the last post seems rather exasperated, it's more a product of that, although I am at a stress plateau with this project.

Put it this way, in simpler terms. If I were to redo the existing amp, it would be a 100% identical rebuild. I'm willing to do that, but at a later time. Much of my frustration came from the enclosure, and so this project (the one which I just posted the circuit for) will be mounted in a old power supply case (if I can reduce the variables I have to work with, I reduce the number of errors I could be making)!

Chris, you were right, I intend for the original build on this thread to be a guitar amplifier. The second one (the schematic I posted last page) is intended to be a headphone amplifier.

And I'm sorry about my quirkiness with all these concepts. I understand them in simple circuits, but when you give me circuits like these, there are so many variables to consider that what I would understand in a perfect-world setting isn't so clear cut for me. All the simple tutorial examples of "battery and light-bulb" I understand perfectly.

Thanks for bearing with me and not swearing at me in your frustration :p
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
If that is a bridge rectifier and two resistors, you are wasting about 97% of the power. A single capacitor and the two LEDs would also work, wasting no power.

Bob
 

Attachments

  • LEDS.jpg
    LEDS.jpg
    13.1 KB · Views: 164
Last edited:

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
Different design, different purpose. Hence the solid state section for secondary amplification.


I have been down the road you are on many years ago. It is very worthwhile to aspire to good distortion sounds. Generally you won’t get it from such a setup as the first schematic you presented. A plate voltage such as 9V is very low and you won’t really achieve a nice tube tone. Even with a ‘proper’ plate voltage and typical values such as in a guitar amp with a typical 12 AX7 it won’t sound that great into your solid state amp and hifi type speakers. That said, I would not want to discourage you from building such a cct. It will be ultimately useful if you build it in a stand alone box as one day you may find it adds a sound you are after. The learning from making a tube preamp will help when attempting a full tube amp with 400 volts or more B+. Also, this cct has very low gain. It will be best with a 12AX7. Even then, it will only add a little bit of crunch, never a lead tone. Such a preamp is usually called an overdrive. Not because it sounds like an overdriven tube but because it is intended to overdrive the preamp of a tube amp, or provide that bit of extra gain so that the output tubes are driven more into saturation. It will probably sound much the same as an overdriven signal transistor. A popular overdrive setup is the Ibanez tube screamer. Plenty on the web about it and it would be a better overdrive than most. It still won’t give you a tube tone when overdriving a solid state amp, but can still be useful when set right to approximate typical commercial tones. It is worthwhile making as when you ultimately buy/build a tube amp it will be most useful.

The schematic I saw today with 3 tubes and 2 mosfets is bizarre. The sources of the mosfets should not be grounded, they will likely destroy themselves in an instant since trhe load is zero. Even then the fets look redundant as they share an input. The tubes are paralled, this is a waste. It looks like some attempt at a stereo to mono converter, but it sure looks useless as a guitar preamp. Would be a disappointment.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
I have been down the road you are on many years ago. It is very worthwhile to aspire to good distortion sounds. Generally you won’t get it from such a setup as the first schematic you presented. A plate voltage such as 9V is very low and you won’t really achieve a nice tube tone. Even with a ‘proper’ plate voltage and typical values such as in a guitar amp with a typical 12 AX7 it won’t sound that great into your solid state amp and hifi type speakers. That said, I would not want to discourage you from building such a cct. It will be ultimately useful if you build it in a stand alone box as one day you may find it adds a sound you are after. The learning from making a tube preamp will help when attempting a full tube amp with 400 volts or more B+. Also, this cct has very low gain. It will be best with a 12AX7. Even then, it will only add a little bit of crunch, never a lead tone. Such a preamp is usually called an overdrive. Not because it sounds like an overdriven tube but because it is intended to overdrive the preamp of a tube amp, or provide that bit of extra gain so that the output tubes are driven more into saturation. It will probably sound much the same as an overdriven signal transistor. A popular overdrive setup is the Ibanez tube screamer. Plenty on the web about it and it would be a better overdrive than most. It still won’t give you a tube tone when overdriving a solid state amp, but can still be useful when set right to approximate typical commercial tones. It is worthwhile making as when you ultimately buy/build a tube amp it will be most useful.

The schematic I saw today with 3 tubes and 2 mosfets is bizarre. The sources of the mosfets should not be grounded, they will likely destroy themselves in an instant since trhe load is zero. Even then the fets look redundant as they share an input. The tubes are paralled, this is a waste. It looks like some attempt at a stereo to mono converter, but it sure looks useless as a guitar preamp. Would be a disappointment.

Very interesting.

What would a more appropriate plate voltage be for the sound?

Also, about the schematic you saw me post on the last page or so - if you read through this thread, you'll get the idea real fast that I don't have much experience with this at all, and the build that I'm working through here is the learning point through which I'm gaining an idea about what electronics means and actually does.

That schematic I drew up to see if I could get the gist of this electronics amplification and what not. I will admit I threw the third tube in there so I will have a third tube lighting up, not necessary for the function I was looking for.

It is designed to be a headphone amplifier rather than a guitar preamp.

If I were to cut out the shared input to the FETs (have each one going to only one tube) and cut out the third tube, where would I place the source of the MOSFETs?
 

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
Very interesting.

What would a more appropriate plate voltage be for the sound?

Also, about the schematic you saw me post on the last page or so - if you read through this thread, you'll get the idea real fast that I don't have much experience with this at all, and the build that I'm working through here is the learning point through which I'm gaining an idea about what electronics means and actually does.

That schematic I drew up to see if I could get the gist of this electronics amplification and what not. I will admit I threw the third tube in there so I will have a third tube lighting up, not necessary for the function I was looking for.

It is designed to be a headphone amplifier rather than a guitar preamp.

If I were to cut out the shared input to the FETs (have each one going to only one tube) and cut out the third tube, where would I place the source of the MOSFETs?

200 volts is a typical starting point, most amps use the main B+ at about 400 volts. The data sheet is worth reading in detail.

The headphone amp only needs 1 tube and 1 Mosfet. The input grids should have series resistors at least, as the are shorted AC wise from left to right. Your stereo source won't like that. Configuring the mosfet as a source follower is OK to do. Look at similar ccts on the web, I have seen plenty of tube headphone preamps. I recommend starting with a basic classical cct and adapting it to your use.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
Okay. Is it okay for the drain to be grounded?

Did I just have the symbol flipped the wrong way?

I intended it to be that the source is supplied by the 12V B+ rail, and the drain is grounded and connected to the jack.
 

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
Okay. Is it okay for the drain to be grounded?

Did I just have the symbol flipped the wrong way?

I intended it to be that the source is supplied by the 12V B+ rail, and the drain is grounded and connected to the jack.

No. It is N-type so the drain must be more +ve. ie to +12V with source via resistor to ground. Even with this config you would need high impedance headphones. The 100k output pot will not work very well. I don't have time right now to look up a good basic design.
 

Solidus

Jun 19, 2011
349
Joined
Jun 19, 2011
Messages
349
Oh okay. Just let me know about the design when you have time.

How do I lower the required impedance for the output?

After I find that out, I can do some homework tonight on a better adaptation to circuit.
 

Geoff C

Sep 11, 2012
9
Joined
Sep 11, 2012
Messages
9
Oh okay. Just let me know about the design when you have time.

How do I lower the required impedance for the output?

After I find that out, I can do some homework tonight on a better adaptation to circuit.

Generally you use a push-pull topology for the output stage. This can be discrete or a small power amplifier IC. An inefficient way is to put a lot of idle current thru the fet and reduce the source resistor to a very low value. This will flatten the battery quickly though, and need a big heatsink.
 
Top