Giaime,
Look for your chassis hiding in other objects. A cover for a VCR/DVD could be a chassis; also an aluminum pan for baking cakes... CHEAP!!!
Best Holidays to you and Santa has confirmed your ECC88's!
Look for your chassis hiding in other objects. A cover for a VCR/DVD could be a chassis; also an aluminum pan for baking cakes... CHEAP!!!
Best Holidays to you and Santa has confirmed your ECC88's!
jane said:
Yes, I thought so. But on the board they're wired correctly. Anyway, it doesn't matter, since I decided to avoid the use of PCB's 😉
poobah said:
I've got a friend who mades metal tanks for oil, wine, etc etc... I will ask him! 😀
Thanks a lot for those tubes, I'm looking forward the arrival
Good that you have a friend with sheetmetal shop! Now you need a friend with a transformer shop! - 😀
Couldn't we all use a friend in a transformer shop? Oh happy days! 
Giaime, at least you know someone who can make some really stylish Italian cases for your projects. Where art and sound collide. 😉
-Chris
Giaime, at least you know someone who can make some really stylish Italian cases for your projects. Where art and sound collide. 😉
-Chris
I don't think he can do the "italian style"... he makes oil tanks, and he plays on an extreme-heavy-metal band.... anyway, art is just art 😉
Hello to everyone! I thought I had to up this thread instead of making a new one 😉
I realized that I really need some gain in my line stage. Actually, it's a cathode follower with a 12AT7, but I would like some little gain, at least 3 or more.
I thought of using the 12AU7 (for its low mu) as a grounded cathode stage DC coupled to the grid of the 12AT7 cathode follower.
The problem is, obviously, to determine the values of the various resistors needed.
So I purchased TubeCAD: I tried it and found the way to design a grounded cathode/CF direct coupled, but you can calculate it with just one tube type. I mean, the tube of the grounded cathode stage must be the same of the CF stage: instead, I'd like to use a 12AU7 for the grounded cathode and a 12AT7 for the cathode follower.
How should I proceed?
1) draw a generic grounded cathode circuit with the 12AU7 (schematic included below) and then, given the plate voltage (the same dc voltage of the CF grid), I can calculate the CF cathode resistor value (how?)
2) stick on using one tube for both the CF and the grounded cathode stage. This leads to 2 drawbacks:
a) I haven't got 2 12AU7
so I can only do with 2 12AT7
b) with 2 12AT7 I will have possibly differences between the left and right channel because different tubes are used, and the biggest problem is the BIG gain I will get, totally useless for me in a line stage.
Please help me
And another question: will the grounded cathode stage I calculated in TubeCAD accept full 2Vrms input without clipping? How to verify this? I think I have to look at Vbias, or am I wrong? I figure I need about 6V of bias to accept 2Vrms without clipping...
I realized that I really need some gain in my line stage. Actually, it's a cathode follower with a 12AT7, but I would like some little gain, at least 3 or more.
I thought of using the 12AU7 (for its low mu) as a grounded cathode stage DC coupled to the grid of the 12AT7 cathode follower.
The problem is, obviously, to determine the values of the various resistors needed.
So I purchased TubeCAD: I tried it and found the way to design a grounded cathode/CF direct coupled, but you can calculate it with just one tube type. I mean, the tube of the grounded cathode stage must be the same of the CF stage: instead, I'd like to use a 12AU7 for the grounded cathode and a 12AT7 for the cathode follower.
How should I proceed?
1) draw a generic grounded cathode circuit with the 12AU7 (schematic included below) and then, given the plate voltage (the same dc voltage of the CF grid), I can calculate the CF cathode resistor value (how?)
2) stick on using one tube for both the CF and the grounded cathode stage. This leads to 2 drawbacks:
a) I haven't got 2 12AU7
so I can only do with 2 12AT7b) with 2 12AT7 I will have possibly differences between the left and right channel because different tubes are used, and the biggest problem is the BIG gain I will get, totally useless for me in a line stage.
Please help me
An externally hosted image should be here but it was not working when we last tested it.
And another question: will the grounded cathode stage I calculated in TubeCAD accept full 2Vrms input without clipping? How to verify this? I think I have to look at Vbias, or am I wrong? I figure I need about 6V of bias to accept 2Vrms without clipping...
The cathode follower won't change the gain or the operating point of the previous stage. So you can design the voltage amp in isolation. Then when you're happy with that, the CF is a slam dunk, designable on the back of a small envelope.
But do reconsider using a 12AU7 unless you want the preamp to have a sound of its own. As a voltage amp, it has a relatively high third and fifth harmonic distortion. That's useful for guitar amps, not desirable in a hifi amp.
But do reconsider using a 12AU7 unless you want the preamp to have a sound of its own. As a voltage amp, it has a relatively high third and fifth harmonic distortion. That's useful for guitar amps, not desirable in a hifi amp.
SY said:
Thank you so much Stuart, so I'll have to avoid 12AU7. So, is there some possibility to use 12AT7 as both voltage amp and CF? How to reduce that big gain that I get? maybe with a more vertical load line, this implies lower plate resistor, isn't it? But I have another fear: the plate voltage will be around 100-120V, what about the avabile B+ headroom to make a CF? Because the CF will have the grid sitting at 100V, the cathode at some volts more, so only 100V cathode to plate... is this too little to give the CF proper headroom? Or to stay away from non-linear zones?
I will re-design something with the 12AT7 in mind (so now I can use TubeCAD 😉 )
Thanks to everyone!
Here's the problem- you don't want very much gain. Even the 12AU7 gave you far too much. You have several choices:
Change topologies and use feedback to set gain and lower output impedance.
Use just the CF, but use a 2:1 stepup transformer on the input (that will get you 6dB or so of gain).
Use a very low mu tube like 12B4.
Do some really weird biasing and local degeneration to knock down the mu of the voltage amplifier section.
Change topologies and use feedback to set gain and lower output impedance.
Use just the CF, but use a 2:1 stepup transformer on the input (that will get you 6dB or so of gain).
Use a very low mu tube like 12B4.
Do some really weird biasing and local degeneration to knock down the mu of the voltage amplifier section.
What about this? Schematic attached.
I just don't know how to calculate the second stage cathode resistor.
I've got 133V on the grid, so I want the cathode let's say at 140V. How to calculate? I can't just make R = 140/I because I don't know I
Aside from that, first stage looks good. TubeCAD says I've got almost 7x gain, which is too high but I don't care too much, I guess I have to keep the potentiometer on low positions (an anti-log pot should help here, but hard to find...). About -3V bias, which should be good given the max input of 2Vrms.
Any comments?
I just don't know how to calculate the second stage cathode resistor.
I've got 133V on the grid, so I want the cathode let's say at 140V. How to calculate? I can't just make R = 140/I because I don't know I
Aside from that, first stage looks good. TubeCAD says I've got almost 7x gain, which is too high but I don't care too much, I guess I have to keep the potentiometer on low positions (an anti-log pot should help here, but hard to find...). About -3V bias, which should be good given the max input of 2Vrms.
Any comments?
Attachments
That part is easy. The second stage grid voltage is equal to the first stage plate voltage, right? So at any reasonable current, the second stage cathode will have to be just a couple of volts higher.
The grid voltage at the CF is 133V, according to your sim. OK, so the cathode will be at 135. Maybe 136. Maybe 134. Doesn't much matter. Call it 135. We choose a standing current that we want to CF to run. This will be based on what we want it to drive (see the first part of the Heretical article). I'll use 3mA as an example here.
We have 135V on the cathode, we want 3mA, now all we have to do is use Ohm's Law to see that the cathode resistor should be 135/3 = 42K. We can run a little richer and use a 39K or a little leaner and use 45K. Either way, the local feedback ensures that the tube will be running at the chosen current.
Thank you very much Stuart, I figured that out. But the question that was blocking my mind was: if I get a gain of 7x, I'd like to use all this gain (let's hypotize an extremely unsensible power amp), so I have to have headroom. If I put the CF cathode just a couple of volts above the grid, will I have headroom for a 20V signal (about 2Vrms * 1.41 * 7)? Or will the local feedback take care of that?
Thanks so much anyway 😉
Thanks so much anyway 😉
Why in the world would you need a 20V signal? There's not a power amp I've ever seen that needs more than 4 or 5 V, and even that is exceptional. 2V is pretty much standard. Don't design for the unlikely at the expense of the normal.
If you nonetheless want to swing 20V cleanly at a gain of 15-20dB, you need something like a 6SN7. A 12AU7 will be running close to a percent distortion at that level.
If you nonetheless want to swing 20V cleanly at a gain of 15-20dB, you need something like a 6SN7. A 12AU7 will be running close to a percent distortion at that level.
SY said:
Oh yes SY, I understand 😉 I didn't mean that I would like to have 20V at the output (btw I intended 20 as a peak value, not rms), but I would like to have some headroom on normal 2Vrms signals. I think that 2Vrms can be fully handled by the grounded cathode stage, I was just not sure if the CF stage could do the same job.
Thank you so much for your kindness, I appreciate it so much! 😉
Thanks to everyone!
Edit: here's the updated version of the schematics. I'm starting to breadboard it now, I can't wait 😀 It will be hard, I have to fully reconstruct the line stage zone... I will have to remove completely the old CF.
Attachments
A question: now that I'm rebuilding the line stage, I have the chance to redefine some grounds. Where do I connect the ground of the heater supply to the ground of the B+ supply?
Now my grounding scheme is this:
I've got a single point where the earth safety ground connects to the gnd of the rectifier board before the B+ regulator, and to the turntable gnd wire. Do I have to connect the gnd of the heater supply here to this grounding point (thus returning also the B+ gnd here) or I have to connect toghether the two gnd in the "audio" zone, maybe at the inputs?
Now my grounding scheme is this:
I've got a single point where the earth safety ground connects to the gnd of the rectifier board before the B+ regulator, and to the turntable gnd wire. Do I have to connect the gnd of the heater supply here to this grounding point (thus returning also the B+ gnd here) or I have to connect toghether the two gnd in the "audio" zone, maybe at the inputs?
Quote:
Why in the world would you need a 20V signal? There's not a power amp I've ever seen that needs more than 4 or 5 V, and even that is exceptional. 2V is pretty much standard. Don't design for the unlikely at the expense of the normal.
All commercial SS amplifiers are designed to reach full power with an input signal of .775 volts which I believe is zero db. Most home type amplifiers will reach full power with a signal of no more than 2.0 volts. Some tube amplifiers that I have come across have needed more than 2 volts. Mind you I am not an authority on tube amplifier signal needs.
Sy has a good point here. Why design a line stage with output capabilities that will never be used? Up until recently I would have argued that it would be nice to have the gain available. I have learned the hard way that the tradeoff will be increased noise and complexity. In other words you have all that gain and unfortunately you also have the noise associated with it. Maybe a solution might be to incorporate your design with the noise cancelling abilities of the Aikido. If my memory is correct the gain will be cut in half and the noise will be gone also.
Sy, is there a way to incorporate this design with the Aikido's power supply cancelling stage?
Why in the world would you need a 20V signal? There's not a power amp I've ever seen that needs more than 4 or 5 V, and even that is exceptional. 2V is pretty much standard. Don't design for the unlikely at the expense of the normal.
All commercial SS amplifiers are designed to reach full power with an input signal of .775 volts which I believe is zero db. Most home type amplifiers will reach full power with a signal of no more than 2.0 volts. Some tube amplifiers that I have come across have needed more than 2 volts. Mind you I am not an authority on tube amplifier signal needs.
Sy has a good point here. Why design a line stage with output capabilities that will never be used? Up until recently I would have argued that it would be nice to have the gain available. I have learned the hard way that the tradeoff will be increased noise and complexity. In other words you have all that gain and unfortunately you also have the noise associated with it. Maybe a solution might be to incorporate your design with the noise cancelling abilities of the Aikido. If my memory is correct the gain will be cut in half and the noise will be gone also.
Sy, is there a way to incorporate this design with the Aikido's power supply cancelling stage?
The Aikido does add some complexity to achieve noise cancellation. With my preamp, I took a bit more of a brute-force approach. As it happens, I've got an Aikido board and am about halfway through a build, so I'll be able to comment more intelligently in a few weeks.
It still makes sense to me to not build in the excessive gain in the first place. Headroom is just not an issue with a hard 2V limit as long as there's no blocking on overload. And with a direct coupling between stages, there won't be, especially with a much-better-than-6dB overload margin.
It still makes sense to me to not build in the excessive gain in the first place. Headroom is just not an issue with a hard 2V limit as long as there's no blocking on overload. And with a direct coupling between stages, there won't be, especially with a much-better-than-6dB overload margin.
SY
I will await your findings. I believe you will be very pleased with the Aikido's performance. I have completed the first of the sets of Aikido boards I have. I have also built a 6AU6 version that has way too much gain but the sound is to die for.
Anyone, is there a way to incorporate this design with the Aikido's power supply cancelling stage? Maybe this should be another thread.
I will await your findings. I believe you will be very pleased with the Aikido's performance. I have completed the first of the sets of Aikido boards I have. I have also built a 6AU6 version that has way too much gain but the sound is to die for.
Anyone, is there a way to incorporate this design with the Aikido's power supply cancelling stage? Maybe this should be another thread.
- Status
- Not open for further replies.