Replacing Transistor with Tube

[Govmnt_Lacky]

Ha!!!

If anything... I was shaming RG about his choice of ESD-safe carpeting to do his work on in the picture 

That... and EVERYONE knows you should use a rubber mallet for that job! 

[Thecomedian]

Only a steel-head hammer will get you that gritty, broken up tone, though.

[duck_arse]

hatredman? I always read this (so far) as hat-red-man, which is a bit different.

and as for frying? I didn't think there was frying going on. usually when that happens, we all know it's happening.

and it seems to me there is a time limit on threads, after which the "signal" is given, and everyone piles in with stupid comments/puns/jokes/pictures etc. i'ts always fun to try and pick what the signal is, but the thread usually mutates into something else once someone has. picked it, I mean.

[wavley]

".... you're doing a fine job." - wavley

When did I accuse you of doing a fine job?  I must have been drunk that day

Ha!!!

If anything... I was shaming RG about his choice of ESD-safe carpeting to do his work on in the picture 

That... and EVERYONE knows you should use a rubber mallet for that job! 

Only a steel-head hammer will get you that gritty, broken up tone, though.

Yes, but you need a wood mallet if you want that woody/reedy tone.

When I worked at the repair shop for the first year I worked on a card table then they built a new room and built me a brand new bench which they lovingly covered in ozite to stop parts from rolling away, on which I spilled a bottle of flux within a week.  Then there was a permanent sticky spot on my bench.  Within a month I tipped my iron over and burnt a hole in it.

[deparisn]

I wish I could get across in words what I am trying to do but I will give it another go.

The question I am trying to answer is how can I objectively compare the difference in using a tube to compress a signal or using transistors to compress a signal. The issue I am running into is that I only want the tube and the transistor to be a variable and I want the rest of the circuit to remain the same.

The problem I am having is I apparently I just can't take the transistor out and put the tube in. So I am rather stuck. Now if I had someone who maybe modded the BYOC 2 knob compressor and made it tube, then I have just the tube and transistor as changing.

[Johan]

I'm afraid the answer to your question is : you can't, they are too different to work in the same Circuit without anything else changing.
it might seem like they do the same thing, but they do it in totaly different ways..one is a lever the other a Fawcett...
...and..welcome to the forum....ask stupid questions and learn or be ignorant forever...this Place is full of good people/teachers...
J

[wavley]

I wish I could get across in words what I am trying to do but I will give it another go.

The question I am trying to answer is how can I objectively compare the difference in using a tube to compress a signal or using transistors to compress a signal. The issue I am running into is that I only want the tube and the transistor to be a variable and I want the rest of the circuit to remain the same.

The problem I am having is I apparently I just can't take the transistor out and put the tube in. So I am rather stuck. Now if I had someone who maybe modded the BYOC 2 knob compressor and made it tube, then I have just the tube and transistor as changing.

Well, yes and no.

Now, you could use something like a valvecaster as your gain stage and run at the same voltages as a transistor, the supporting circuitry isn't exactly plug and play so you can't just shove a transistor in a tube socket and have it be the only component change, but if you want to just exchange as a gain block it's pretty doable.

OR, you can run at normal tube voltages instead of the starved plate of the Valvecaster and pretty much just drop in a high voltage MOSFET using pretty much the same circuit values, which is what Marshall did in the older Valvestate amps that use a hybrid of tubes and VN2410L MOSFETS, check out the schematic for the Marshall Valvestate 8280 schematic at the Dr.Tube site.   Most of the values are the same with the exception of what the cathode values would be on the tube.

Hope this helps.

edit: and of course you don't need the heater voltages and such for the MOSFET, just a little less circuitry

[Hatredman]

1. ok ok ok I take back what I said.


2. deparisn, the short answer is: what you want is impossible. There is absolutely NO way to use the SAME circuit and only swap a transistor for a tube.


3. One day I'll explain my hatred

[merlinb]

An LND150 will work in the same circuit as a tube...
http://www.redrok.com/MOSFET_LND150_500V_30mA_1kO_Vth-1.0_TO-252.pdf

[wavley]

An LND150 will work in the same circuit as a tube...
http://www.redrok.com/MOSFET_LND150_500V_30mA_1kO_Vth-1.0_TO-252.pdf

Cool, I was only aware on the VN series stuff being able to drop in a tube circuit with very little change.  Can you keep the source values the same as cathode values?  The only time I've seen the VN stuff in circuits the source values were different than the cathode would normally be.

Even so, it's only the difference of one RC combo between the tube and the MOSFET so it's totally doable if you want to build a high voltage gain stage.

[merlinb]

 Can you keep the source values the same as cathode values?  

I would guess 1k or thereabouts would work for both a 12AX7 and LND150, yes.

[wavley]

 Can you keep the source values the same as cathode values?  

I would guess 1k or thereabouts would work for both a 12AX7 and LND150, yes.

Very cool.

And also, there's your answer.  Build a tube circuit and then sub in an LND150.

edit: sub in two LND150s if you plan on using both halves of the tube

[mth5044]

So.. what you're saying is after two pages of jokes about the OP's question, there is an actual transistor that can replace a tube.

[midwayfair]

So.. what you're saying is after two pages of jokes about the OP's question, there is an actual transistor that can replace a tube.

But it's not a 5088 that's being replaced, and has nothing to do with what the op is looking to do.

The transistors used in the BYOC 5-knob compressor aren't gain stages. They are emitter follower stages and one is a variable resistor (control voltage fed to the base).

deparisn, as I explained in detail on your BYOC thread, the 5-knob compressor's compression is done by varying the bias of a CA3080 chip. The chip is your amplifier, and the amplifier part of the chip is an operational amplifier, which involves quite a lot of circuitry. The CA3080's datasheet gives you its internal circuit (on page 3) -- it's 11 transistors and numerous other components. I'm not smart enough to tell you how to translate the internals of that op amp to a tube, and I don't mean to seem rude, but your questions on the forums have indicated that you are probably not advanced enough to do so, and I would be absolutely flabberghasted if anyone has bothered to create a tube-based CA3080 circuit already, although you could look up some very, very VERY old schematics for a tube-based operational amplifier and figure out how to implement all the pins of the chip.

The bias pin on the chip is the source of the compression. A tube by itself would have nothing to do with the compression.

The transistors in the dynacomp circuit aren't part of the sound. The only ones in the audio path are buffers. You can build a cathode follower and an emitter follower and assuming you build the circuits properly, you will end up with a near-unity gain buffered signal in both cases. I have a feeling this won't tell you anything. And I don't think a tube stage can be used as a variable resistor in the way that a transistor can, at least not without huge changes everywhere in the circuit.

If you just want to analyze the difference between a transistor and a tube, you can scope the output of two circuits created to have identical responses: Create them to have identical output and input impedance, and feed them a sine wave signal that is proportionally the same size in relation to the supply voltage the circuit needs to operate.

[deparisn]

 Can you keep the source values the same as cathode values?  

I would guess 1k or thereabouts would work for both a 12AX7 and LND150, yes.

Very cool.

And also, there's your answer.  Build a tube circuit and then sub in an LND150.

edit: sub in two LND150s if you plan on using both halves of the tube

Ok Wait a sec. I am trying to stay away from design as much as possible, what do I need to do or where do I need to go to find a tube compressor circuit run a bunch of tests, and then replace said tube with an LND150 and run the same test and "publish" my findings. I do not mind putting a SPICE circuit that's actually kind of the thing I need I just do not want to design a complete circuit from scratch. My project is one of how we say research not design.

Please people keep telling me things. I have the "what" of my project which is  Study the difference in tube compression and solid state compression. The how is pretty much what I am trying to iron out.

 

[Thecomedian]

I don't know if some of these resources might help.

http://www.theaudioarchive.com/TAA_Resources_Tubes_versus_Solid_State.htm
http://www.ovnilab.com/articles/comptypes.shtml

Tube compressors, almost always, are really optical comps with a simple 12AX7 tube gain stage added on the end. However vari-mu (short for Variable Mu, also a Manley trademark) designs use a vacuum tube in place of a transistor, where the variable voltage input changes a tube's bias instead of a transistor's gain. It has a somewhat different action, generally claimed to be even smoother and more "creamy" or "organic". Aside from the Manley and some very old Fairchild units, the only other vari-mu-style examples I know of are the Markbass Compressore and the Lightning Boy Opti-Mu Prime.

http://www.dwfearn.com/tubes_vs_transistors.html
http://www.effectrode.com/signal-tubes/vacuum-tubes-and-transistors-compared/
http://boards.straightdope.com/sdmb/showthread.php?t=513411
http://www.freeinfosociety.com/media/images/5377.jpg

[Blitz Krieg]

this is all being done in Spice?

[wavley]

So.. what you're saying is after two pages of jokes about the OP's question, there is an actual transistor that can replace a tube.

Yes, there is a high voltage MOSFET that can replace a tube as an amplification stage.  There were a few jokes to see what his application was to get there and there's a pretty big difference between a 9v 5088 stage and a full voltage tube circuit.  So the real answer is, no there is no tube that can replace a 5088, but there is a work around IF it's an AMPLIFIER you're replacing.  BUT, it turns out that even after making jokes to get more information, we weren't given all the information... we had to know about another thread.

So.. what you're saying is after two pages of jokes about the OP's question, there is an actual transistor that can replace a tube.

But it's not a 5088 that's being replaced, and has nothing to do with what the op is looking to do.

The transistors used in the BYOC 5-knob compressor aren't gain stages. They are emitter follower stages and one is a variable resistor (control voltage fed to the base).

deparisn, as I explained in detail on your BYOC thread, the 5-knob compressor's compression is done by varying the bias of a CA3080 chip. The chip is your amplifier, and the amplifier part of the chip is an operational amplifier, which involves quite a lot of circuitry. The CA3080's datasheet gives you its internal circuit (on page 3) -- it's 11 transistors and numerous other components. I'm not smart enough to tell you how to translate the internals of that op amp to a tube, and I don't mean to seem rude, but your questions on the forums have indicated that you are probably not advanced enough to do so, and I would be absolutely flabberghasted if anyone has bothered to create a tube-based CA3080 circuit already, although you could look up some very, very VERY old schematics for a tube-based operational amplifier and figure out how to implement all the pins of the chip.

The bias pin on the chip is the source of the compression. A tube by itself would have nothing to do with the compression.

The transistors in the dynacomp circuit aren't part of the sound. The only ones in the audio path are buffers. You can build a cathode follower and an emitter follower and assuming you build the circuits properly, you will end up with a near-unity gain buffered signal in both cases. I have a feeling this won't tell you anything. And I don't think a tube stage can be used as a variable resistor in the way that a transistor can, at least not without huge changes everywhere in the circuit.

If you just want to analyze the difference between a transistor and a tube, you can scope the output of two circuits created to have identical responses: Create them to have identical output and input impedance, and feed them a sine wave signal that is proportionally the same size in relation to the supply voltage the circuit needs to operate.

So the answer is still no, because we just ran with the assumption that it was an audio gain stage.  Building a high voltage dyna comp and swapping in MOSFETs and tubes isn't doable.  

I for one am not psychic and didn't read or know about the other thread.

Yes, he mentioned compression, but when engineers speak of compression (which we did have that information) they aren't speaking of a compressor like a dynacomp, they're speaking of the onset of clipping where the tops and bottoms of waveforms start being squashed and you have a drop in output power. http://www.microwaves101.com/microwave-encyclopedia/320-compression-point

So it is perfectly reasonable for me to, once I've been given the parameters by an engineer, that they want to test the compression characteristics between tubes and transistors to say "well, you can test the differences between tubes and MOSFETS in the same circuit, maybe you should adjust your test situation"  Of course, no you can't replace a CA3080 with a tube, it's pointlessly complicated with too many variables to be a decent test.

Of course as a microwave R&D technician, dragging the the whole story out of PhD engineers is what I do for a living, so I guess I should have known there was more to the situation than what was being stated, there's always a piece of information somewhere else that you aren't being told about.

This thread has taken a pretty ugly turn with the shaming of folks that are just trying to help somebody out and drag some information out of them.  

edit: Perhaps there should be an entire class added to the engineering curriculum teaching engineers how to communicate all of their needs to technicians and each other.  Not as a slam to the OP, but engineers in general, they seem to be pretty bad at that sometimes.

[mth5044]

I didn't pick up anyone shaming anyone in this thread, I just thought it was funny that someone came up with a tube/transistor swap that got somewhere closer to what the OP had said. While some of the comments are totally unnecessary and may have wounded the OP a bit, the OP was very vague and on a forum where details are key, you can't get very far with vague questions. There was some hint of calling the question stupid, which another poster picked up on, but I don't think anyone was 'shamed' for their comments.

The thread turned ugly after the fourth comment, not whenever this shaming occurred. I didn't mean any offense with my post, just as I'm sure whoever said 'quote of the century' didn't, but it's hard to pick up the meaning behind typed words and easy to misinterpret.

Anyway, I would have thought the OP was going for something like the fetzer from ROG where a triode tube is emulated with a JFET.

http://www.runoffgroove.com/fetzervalve.html

[Thecomedian]

So.. what you're saying is after two pages of jokes about the OP's question, there is an actual transistor that can replace a tube.

Yes, there is a high voltage MOSFET that can replace a tube as an amplification stage.  There were a few jokes to see what his application was to get there and there's a pretty big difference between a 9v 5088 stage and a full voltage tube circuit.  So the real answer is, no there is no tube that can replace a 5088, but there is a work around IF it's an AMPLIFIER you're replacing.  BUT, it turns out that even after making jokes to get more information, we weren't given all the information... we had to know about another thread.

So.. what you're saying is after two pages of jokes about the OP's question, there is an actual transistor that can replace a tube.

But it's not a 5088 that's being replaced, and has nothing to do with what the op is looking to do.

The transistors used in the BYOC 5-knob compressor aren't gain stages. They are emitter follower stages and one is a variable resistor (control voltage fed to the base).

deparisn, as I explained in detail on your BYOC thread, the 5-knob compressor's compression is done by varying the bias of a CA3080 chip. The chip is your amplifier, and the amplifier part of the chip is an operational amplifier, which involves quite a lot of circuitry. The CA3080's datasheet gives you its internal circuit (on page 3) -- it's 11 transistors and numerous other components. I'm not smart enough to tell you how to translate the internals of that op amp to a tube, and I don't mean to seem rude, but your questions on the forums have indicated that you are probably not advanced enough to do so, and I would be absolutely flabberghasted if anyone has bothered to create a tube-based CA3080 circuit already, although you could look up some very, very VERY old schematics for a tube-based operational amplifier and figure out how to implement all the pins of the chip.

The bias pin on the chip is the source of the compression. A tube by itself would have nothing to do with the compression.

The transistors in the dynacomp circuit aren't part of the sound. The only ones in the audio path are buffers. You can build a cathode follower and an emitter follower and assuming you build the circuits properly, you will end up with a near-unity gain buffered signal in both cases. I have a feeling this won't tell you anything. And I don't think a tube stage can be used as a variable resistor in the way that a transistor can, at least not without huge changes everywhere in the circuit.

If you just want to analyze the difference between a transistor and a tube, you can scope the output of two circuits created to have identical responses: Create them to have identical output and input impedance, and feed them a sine wave signal that is proportionally the same size in relation to the supply voltage the circuit needs to operate.

So the answer is still no, because we just ran with the assumption that it was an audio gain stage.  Building a high voltage dyna comp and swapping in MOSFETs and tubes isn't doable.  

I for one am not psychic and didn't read or know about the other thread.

Yes, he mentioned compression, but when engineers speak of compression (which we did have that information) they aren't speaking of a compressor like a dynacomp, they're speaking of the onset of clipping where the tops and bottoms of waveforms start being squashed and you have a drop in output power. http://www.microwaves101.com/microwave-encyclopedia/320-compression-point

So it is perfectly reasonable for me to, once I've been given the parameters by an engineer, that they want to test the compression characteristics between tubes and transistors to say "well, you can test the differences between tubes and MOSFETS in the same circuit, maybe you should adjust your test situation"  Of course, no you can't replace a CA3080 with a tube, it's pointlessly complicated with too many variables to be a decent test.

Of course as a microwave R&D technician, dragging the the whole story out of PhD engineers is what I do for a living, so I guess I should have known there was more to the situation than what was being stated, there's always a piece of information somewhere else that you aren't being told about.

This thread has taken a pretty ugly turn with the shaming of folks that are just trying to help somebody out and drag some information out of them.  

edit: Perhaps there should be an entire class added to the engineering curriculum teaching engineers how to communicate all of their needs to technicians and each other.  Not as a slam to the OP, but engineers in general, they seem to be pretty bad at that sometimes.

As far as OP goes, it seems to me like it'd be relatively easy to change out a stage for even a low power transistor like 2n3904 of 2n5088 with a mini tube that wants a near-equivalent grid input voltage that's coming from guitar or a previous stage, then simply add another "battery" of much higher voltage (1950's) or an HV line (contemporary) along with the proper biasing network, and bob's your uncle. I would think similar would be true even in an actual compressor pedal as long as you can keep each stage from being affects, be it emitter follower feeding into another transistor which affects the OP amp bias, or an actual pre or post gain stage, etc. Circuits seem to me to be very plug and play and modular in concept, at least. Then again I am no EE.

As far as EE's failing to give all the pertinent information, I find that a bit surprising. It seems like a field which would draw the personality type that loves explaining every little thing in detail even when others have lost interest   Every little detail really matters to some types of people, which sounds like it meshes perfectly with EE and would strongly attract them. I know people like that personally, not meaning to offend anyone.

If he's just looking into distortion/compression from overdriving, there's volumes and volumes on the subject in research about how it differs.