|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
206.228.220.38
Looking for some Tube DIY advice. A while back I modified my Supratek Chenin preamp to run 2C22/7193 tubes in place of the 6SN7, by fabricating adapters to handle the differing pinouts. I also moved the output wires to triode 1 since the 2C22 is a single triode. The sound is much cleaner than stock, and clearer, and I like it a lot, and the lower gain gives me more range on my volume control.
I did not change any of the voltages at that time because I wasn't sure if I'd like the change to the 2C22 and I didn't want to hack the preamp too badly in case I wanted to go back. I'm now positive that I'll never go back to 6SN7s, and I'm thinking I can probably optimize the sound a bit more by adjusting the 2C22 parameters. Looking at the plate curves for the 6SN7 versus the 2C22, it is apparent that the 2C22 will probably like a bit more plate voltage than the 6SN7 - the most linear part of the curve of the 2C22 is more toward the upper end of the voltage ranges than the 6SN7. For the 2C22 with grid voltage = 0, the linear range is from 120 - 175V, whereas for the 6SN7 it is from 70 - 140V. The Chenin plate voltage (measured with tubes in place) is ~100V, perfect for the 6SN7, a bit low for the 2C22.
So my question is, since the Chenin is designed to be self-biasing like most preamps, can I simply change the plate resistor value to increase the plate voltage from 100V to ~150V, or will I need to play with the cathode resistor (and thus the bias) as well?.
I was thinking of temporarily putting a pot in place of the plate resistor, dialing in what I like best, then replacing the pot with a fixed resistor of the same value. I don't know if the "self-biasing" can handle a change from 100V to 150V plate voltage, however, hence the question.
Any help is greatly appreciated!
Well, I measured the cathode voltage and it's 3V, so the grid voltage referenced to the cathode is -3V. Given that and the plate curves for the 2C22 I'm thinking 175 plate volts with -3V on the grid looks just about right - gives 16mA plate current and 2.8W plate dissipation, where the 2C22 max is 3.3W.
Puts me smack in the middle of the linear region for the 2C22 as well. 100V is in a non-linear region, so I'm surprised it sounds as good as it does; can't wait to hear it in the linear region.Thanks for all your help with this!!! I think I've got it now...
Edits: 07/01/09
I would have never used a 6SN7GT at 100 plate volts. A 6P5GT or 76? Yes and for preamp use only. Never a driver or phase splitter.
Max
I have done a couple of the Suprateks now and there is the voltage adjustment in most of the units which you adjust to 300vdc on the source for the 6L6gc thats in there..Do you have the 6L6gc in the preamp before you make changes?
If the power supply waveform isn't pretty,neither is the sound in most cases.
Mike, there are two pots in the unit near the ouput trannies. Are those the adustments you are talking about?
I am currently running a GZ33 as the rectifier and the New Sovtek KT-66s as regulators, but i still have the original tubes if I need to put those back in.
Those be the one max..You set them for 300vdc.Also,don't use the kt66s there..You can but its silly and I would use the 6n3c-e that we talk about as thats what was in it originally,at least the ones I got in..
If the power supply waveform isn't pretty,neither is the sound in most cases.
Edits: 07/01/09
Mine came with Sovtek "5881" as regulators, which are not really 5881s at all, but 6l6GCs. I tried NOS Tung-Sol 5881s there and thought it was an improvement, then I went with the Sovtek KT-66s (as recommended by Mick, the designer) and they sounded better than either of the others, with fuller sound and more overall sonic impact.
Thanks for the info on the adustment pots. I'll make sure those are dialed-in correctly before I start on anything else.
Oh and to answer your question, Mike, yes all the tubes are in and warmed up when I make the measurements. It's really easy to measure with the 2C22 tubes since the plate and grid are top caps - just slip the probe under the edge of the cap with the unit on, other probe on the chassis.
I'm totally with you one the 2C22 sound . I using them on my preamp OPT coupled and it's up there with any expensive preamp .
however if I understand your post correctly that you running only one per chanel VS one 6SN7 per chanel right ?
LT
Yes. The Supratek uses one 6SN7 per channel, with triode I anode directly coupled to triode II grid to give a whole lot of gain. I didn't need two gain stages, so I made up some adapters so the 2C22 fit the 6SN7-wired socket, and I moved the two ouput wires from triode II to triode I on each side - that way I only have one gain stage per side. It's plenty of gain and I now have full travel on my volume control.
Maxamillion, I have no experience with 2C22 tubes, however the RCA data sheet recommended operating point is -10.5V grid (or + 10.5 V cathode bias) at 300V plate voltage and 11 mA plate current.
You will find it here:
http://www.mif.pg.gda.pl/homepages/frank/sheets/049/2/2C22.pdf
Take a look at the -10.5 line on the last graph.
Most 6SN7s run around 9V cathode bias and 180-250 V plate voltage, so you are correct in thinking that a different op point may be better, however, it is dangerous to look for the best op point around 0 Volts grid. The proper way to "ballpark" a good op point is to look for equally spaced grid lines around a SAFE op point. Equal spacing shows linearity, meaning that for a given grid input signal, the plate signal will swing equally in the positive and negative portion of the waveform.
By SAFE I mean: most triodes will go into thermal runoff (lack of control, maximum plate current, red plates, kaput) near 0 Volts, so a good starting point is the manufacturer's recommended op point.
You may work up or down from there, but anything lower than 3 V bias is asking for trouble. You want to stay away from 0 volts even at maximum volume with a highly modulated CD (2V RMS).
Also do not exceed the maximum power of 3.3 Watts.
300V * 11 mA = 3.3 Watts, so you may want to reduce the plate voltage or plate current by 20% to be on the safe side.
I suggest that you post the power supply (B+) voltage, a simple schematic and ask for favorite operating points from those that have experimented with 2C22 tubes.
A change in plate resistor will certainly require a change in cathode resistor to bring the tube to your chosen operating point.
Good luck
I have used 6sn7's at 210V, and 15mA with a CCS for about 4V on the grid. They sounded great. The pre I was using at the time put out about 10V max. You also need to consider the amp's input requirement. There is no definite setting for voltages without a necessary input / output voltage requirement. You can even drive a 300B with a 6sn7 with the right approach.
I see your point on plate dissipation, Casouza; it looks like 150V is too much at a grid voltage of 0. Somewhere around 125-130V keeps the dissipation at ~ 70% of max.
Thanks!
Maxamillion, your grid voltage may be zero Volts, however tubes are biased by the difference between grid potential and cathode potential. Usually, the cathode is run at a positive voltage by means of a cathode resistor.
I suggest that you locate the cathode resistor and measure the voltage across it.
That's your actual grid-to-cathode bias voltage.
Divide the voltage by the cathode resistor value in ohms, you will find out the current in miliamperes.
For example on a typical 6SN7 circuit you may find 9 volts across a 1500 ohm cathode resistor:
9V/1500 ohms= 0.006A = 6 miliAmps
I suggest that you read this to clarify my point:
http://www.bonavolta.ch/hobby/en/audio/t_bas9.htm
Yes, that's what I was saying. The grid voltage is 0V, not the bias. I don't see any magic about 0 or any other absolute voltage, it's all relative, so as long as the grid is negative compared to the cathode, and both are negative relative to the plate, what's the difference where the grid is set? Am I off base here?
Tubes are simple analog devices. All curves, etc, are informaion about events from the tube's perspective.
Tubes do not realize anything is grounded.
Measuring relative to ground is only done to try to simplify documentation.
It might ... or might not ;)
Happy Ears!
Al
I have the datasheets for both the 2C22 and 6SN7. The operating points are exactly where Supratek set them for the stock tubes (one 6SN7 per side) - the first half of the 6SN7 is run at 0V grid and 100V plate and the plate is directly coupled to the grid of the second half of the 6SN7, which is run with the plate at 230V. This is typical of what is shown in the 6SN7 datasheet for a two stage amplifier. What I did was to substitute the single triode 2C22 for the first half of the 6SN7, and just go without a second gain stage, so the anode and cathode of the 2C22 are connected to the output tranformers through a DC blocking cap. The 2C22 thus sees the same voltages that the first half of the 6SN7 did
It sounds excellent like this, believe it or not! Looking at the 2C22 plate curves on the datasheet, however, it looks like going up to 150V or so will put me in a more linear part of the curve, so I'd like to try that, hence the post.
When I change plate voltages, I then measure the change in idle bias current of tube (measure drop across cathode resistor & calculate the current). Normally, if the voltage change isn't too great on the anode plate, the idle current will compensate somewhat and you'll still be in an appropriate idle current operating range. If not, then start playing with the cathode resistor. I measure with no inputs connected and zero volume (if you have control).
Thanks, that's the info I was looking for! So I will monitor the idle current and adjust the cathode resistor value if needed.
Have you tried using a pot (rheostat) on the plate to adjust the voltage? I know the guitar guys like to do that sometimes. If so, what pot did you use (wattage, ohms, etc.).
Yes, I have on output tubes, when trying to dial-in Class "A" cathode biased amps. I usually put a 100 to 150-ohm rheostat inline with a fixed resistor, say 100 ohm, while fine tuning the bias to tone (eg: for an amp that calls for a 220-ohm resistor). The fixed resistor in there just in case the rheostat fails or someone (read: me) cranks the thing the wrong way.
Once I'm happy with tone and status of tube, I replace rheostat and fixed resistor with one unit. Check link for rheostats.
As for gain stage tubes, I haven't used a potentiometer. I play the adding parallel and/or series combos until I hit the point I like (I try for midway between saturation and cut-off). Then, measure the resistance of the mess and replace with one resistor of near equal value. I don't see why you couldn't use a pot, though.
Thanks for the info, FL!
Post a Followup:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2009, Audio Asylum, all rights reserved ]
[ Go Green ]
[ Take5 ]