-10 unbal to +4 xfrmr balance how much gain?

[Blue Jinn]

But 6418 can't handle a +4dBu signal level - my circuit only does +0dBV (+2dBu) as a maximum level. For +4dBu, you really need about a +22dBu max level. Even if you step up the plate voltage to 30V, you aren't going to get there. And since plate current is limited to 500uA, you can't use a step-up transformer to do it either.

Thus, you must follow it with a transistor(s) that is doing all of the work--voltage and current gain. At that point 6418 is just a harmonic distortion generator, and a noisy one at that. Replace V1 with another J107, with a common source and 10K drain resistor, and you've got pretty much the same thing.

Obviously I'm missing something. My understanding is "consumer level" -10dBV = about .775 volts (for "0VU") and "pro level" +4dBu = about 1.23 volts for "0VU"

The 6418 data is referenced to a .85 volt signal which would be about the nominal 0VU -10 dBV signal. The data indicates a max grid input of 1.2 volts (albeit for 12% distortion...) If the max signal going in is going to be nominally .775 volts (consumer level 0VU) and the gain is going out is to be about 4-8x (the tube providing some and the transistor the rest) I'm not clear on the problem unless it means unacceptable distortion if you hit the input with >.775 volts.

 

[mshilarious]

Your references are off a bit:

0dBV = 1V
-10dBV = 316mV

0dBu = 775mV
+4dBu = 1.233V

0VU in either case is nominal level, not maximum level. Generally, you would want at least 15dB of headroom above that. +19dBu is 7.1V. That's an RMS measure; peak-to-peak is 2.828 times that, or 20V.

Let's go back to 6418. My circuit had an 18V supply, the tube actually had gain of 4 (I cheated and took gain of 2 from the FET). My 0dBV max figure was at a lot less than 12% THD, which is going to be unacceptable for anything but a guitar amp. Before you think, hey, second order distortion is a great thing, I used the maximum second order before higher order distortion started to rise as my cutoff. But even if you accept that, realize that you never get pure harmonic distortion without intermodulation distortion, which even if you could accept 12% THD, you couldn't accept the IM you'd get with it.

Let's say a 30V plate, now you have tube gain of 6 rather than 4, and thus +4dBV, maybe +6dBV if you aren't fussy. That's fine for -10dBV level, still not adequate for +4dBu.

Remember that normal peak-to-RMS factors for music are around 15dB, but for sine waves that figure is only 3dB. Therefore, you can't treat 0dBV program material like a 0dBV sine wave. You need headroom.

Also, note that getting gate bias correct is absolutely critical for low distortion, 0.1V on either side of ideal (for that particular 6418) increased distortion, reduced gain, or both, rather significantly.

 

[Blue Jinn]

Your references are off a bit:

0dBV = 1V
-10dBV = 316mV

0dBu = 775mV
+4dBu = 1.233V

You are of course correct. I was misremembering/confusing 0VU with some other reference. (Probably the dB scale on my cheap little multimeter...)

0VU in either case is nominal level, not maximum level. Generally, you would want at least 15dB of headroom above that. +19dBu is 7.1V. That's an RMS measure; peak-to-peak is 2.828 times that, or 20V.

Here is where I'm not understanding this. Take for example my board: has a nominal out of 0VU=-10dBV=.316V. The peak light is calibrated at 0dB or ~1volt. I'm intending this to be used with a mxidown deck and most likely after a limiter as well, so the levels should be fairly consistent and below 0VU. With 15dB of headroom, even if I go 15dB over 0VU, which I'm hoping to not do, aren't I still only at 1.7volts at the input? With 8x gain after the tube and transistor, I'm at 14.15 volts. Pretty close to the power supply, but if I take that up to 30v don't I have enough room there? With the tube giving me 4x gain the output from the tube is about 7 volts, within the psu limits. (Again I'm not clear what I'm missing.)

If I'm generally keeping my levels in the black and I'm rarely going to hit that 1 volt peak I don't see where I'm driving the 6418 that much harder even at 0VU.

Is the issue that after about -18dBV there just isn't any where to go after that, (too much distortion) or am I missing something altogether? If all I need to do is take <= .316v and make it <= 1.234 volts (a 4x gain) I'm not following the drawback here. Or as you indicated later, is it that 0VU is just a mean value and there is a lot more going on there (about 15dB worth)?

 

[mshilarious]

Here is where I'm not understanding this. Take for example my board: has a nominal out of 0VU=-10dBV=.316V. The peak light is calibrated at 0dB or ~1volt. I'm intending this to be used with a mxidown deck and most likely after a limiter as well, so the levels should be fairly consistent and below 0VU.

OK, so you have an average level of -10dBV, and peak maybe a little over 0dBV.

With 15dB of headroom, even if I go 15dB over 0VU, which I'm hoping to not do, aren't I still only at 1.7volts at the input?

Yes.

With 8x gain after the tube and transistor, I'm at 14.15 volts. Pretty close to the power supply, but if I take that up to 30v don't I have enough room there? With the tube giving me 4x gain the output from the tube is about 7 volts, within the psu limits. (Again I'm not clear what I'm missing.)

No. You are talking RMS voltage, not peak-to-peak voltage. Vpp is 2.828x RMS. So 7.1VRMS requires a 20V supply AND an amplifier that can swing within that range with low distortion. That's a piece of cake for an opamp, a bit more difficult but still possible with a single FET (with a bit of feedback), impossible for 6418. Feedback would help a bit there too, I would guess, but you'd lose a bit of gain, and you aren't going to achieve 7VRMS no matter what you do.

Is the issue that after about -18dBV there just isn't any where to go after that, (too much distortion)

Yes. Essentially, the RMS voltage increases because the output starts to resemble a square wave. A square wave has RMS = peak (or 1/2Vpp). So the extra "gain" is all distortion, practically speaking.

At some point you should just breadboard the circuit and play around with it, it isn't that many components. I've got to do a crazy current-sharing dual opamp headphone amp first, then maybe I'll try to see what 30V and a few more components can do . . .

 

[Blue Jinn]

No. You are talking RMS voltage, not peak-to-peak voltage. Vpp is 2.828x RMS. So 7.1VRMS requires a 20V supply AND an amplifier that can swing within that range with low distortion. That's a piece of cake for an opamp, a bit more difficult but still possible with a single FET (with a bit of feedback), impossible for 6418. Feedback would help a bit there too, I would guess, but you'd lose a bit of gain, and you aren't going to achieve 7VRMS no matter what you do.

OK, THANKS, now I think I get it.

Yes. Essentially, the RMS voltage increases because the output starts to resemble a square wave. A square wave has RMS = peak (or 1/2Vpp). So the extra "gain" is all distortion, practically speaking.

So IOW the tube has to be able to handle the voltage swing for the given amount of gain, otherwise it gets clipped, right?

At some point you should just breadboard the circuit and play around with it, it isn't that many components. I've got to do a crazy current-sharing dual opamp headphone amp first, then maybe I'll try to see what 30V and a few more components can do . . .

Why not. (First I need to get the bench cleaned off... )