Line level into a 5751 pre?

[shaun]

Stoopid question of the week: Will a line level signal (which I believe to output at 2-3V), running into a 5751 preamp tube in V1 position, burn out the 5751? I believe a 5751 would be happy with a typical guitar input signal - between 50mV and 750mV.

But me, I connected a turntable with a line level output to a tube amp with a 5751 in V1 as follows: Rk = 1.5kohms; Ra = 100kohms. Plate voltage right around 300VDC.

The 5751 failed, and the Rk looks like it's been overheated. I'm guessing it was because the line level signal drove the 5751 beyond its happy place. In which case, I'd need an L-pad. Or a whole new game-plan.

Why would I plug a turntable into a tube amp? It sounded good. Hyuk. Albeit in mono.

[DummyLoad]

5751 won't be happy. pad the input; e.g., voltage divider.

hint: think mixing board. 


--pete

[pdf64]

Wattage rating of the 1k5 ‘overheated’ cathode resistor?
Consider that a 1/2W would need over 27V cathode voltage for that to happen.
So, if there’s a cathode voltage problem, I don’t see how it could be signal level related here.

[shaun]

Thanks pdf64. It was a half watt. I replaced it with a 1w. I guess it's time to teach myself the nuances of signal current. I can build a decent amp, finally, but I still only have half an idea of the details. I suppose it takes many years to be really good at it all.

Yes, Dummyload. Thanks for that. I googled the voltage divider aspect, and it seems an L-pad is the simplest fix...if I can only figure out the math...

I'd probably want to drop the signal voltage in by about two thirds.

Vout = Vin * R2 divided by (R1 + R2), so I'll have to remember my high school algebra to get the values required. This should be an interesting exercise.

[brewdude]

Volume pot before the input grid? 

[labb]

Think Garnet Amps

[shaun]

Volume pot before the input grid?

Good idea. I would if there was room on the amp for one more knob. Doh!

I came across these L-pad numbers for a voltage divider where R1=4.7ohms and R2=1ohm:

3V (line-level in to pad) * 1ohm (R2) divided by [4.7ohms (R1) + 1ohm (R2)]

3 * 1 divided by 5.7 = 0.526 of a volt output. Which looks about right and should keep my pre tubes happy. Gotta love math. At least when it's this simple 

[2deaf]

Your tube may be happier with 0.526 of a volt, but your line level device isn't going to be happy driving 5.7Ohms.

[shaun]

Your tube may be happier with 0.526 of a volt, but your line level device isn't going to be happy driving 5.7Ohms.

Hmmm. I assumed it would also be driving the grid leak and tube R in parallel, the sum of which would be determined by the value of the grid leak R.

So, if that's not the case, then I'd need to increase the R values in the voltage divider like so?

3V * 47k divided by (220k + 47k) = 0.528V

This way, the line level output would be driving 267Kohms. Which I think would be much better. But I'm guessing, and just trying to connect the dots.

Thanks for your help, btw.

[sluckey]

3V * 47k divided by (220k + 47k) = 0.528V

Much better.

[pdf64]

...I came across these L-pad numbers for a voltage divider where R1=4.7ohms and R2=1ohm...

Hmmm. I assumed it would also be driving the grid leak and tube R in parallel, the sum of which would be determined by the value of the grid leak R...

So you were looking at a speaker level L pad, and that's how you got such low resistances?
The grid leak limit for 12AX7 is typically around 2M.
What did you mean by 'tube R'?
Note that unless it's being overdriven (or operated in 'grid current' mode), a tube control grid's input resistance may be regarded as being effectively infinite, its equivalent circuit is just its Miller capacitance.

Line level input impedance is typically in the range 10k - 100k. There's no downside per se in going higher but for good audio fidelity, potential divider resistances are typically kept below 100k. As the values used increase beyond that, parasitic / stray capacitances will tend to interact with the resistances to create unintended high and low pass filters, thereby messing up the flat 20-20kHz frequency response that so much effort goes into achieving.
This looks to be a good article about the interconnect arrangements used in audio systems https://www.soundonsound.com/techniques/understanding-impedance

[shaun]

Thank you pdf64! I will take a good look at that info. Much appreciated.

[DummyLoad]

your tube is very happy with a 1/2V drive but you have 3V, so you want a 6:1 divider with about 50K input load: 47K/8.2K or 56K/9.1K would be close enough.


--pete 

[shaun]

[/quote]
So you were looking at a speaker level L pad, and that's how you got such low resistances?
[/quote]

Yes, it must have been for a speaker. Good call.

[shaun]

your tube is very happy with a 1/2V drive but you have 3V, so you want a 6:1 divider with about 50K input load: 47K/8.2K or 56K/9.1K would be close enough.


--pete

Thanks Pete. I think I'll go with the 47K/8.2K combo.

[shaun]

This looks to be a good article about the interconnect arrangements used in audio systems https://www.soundonsound.com/techniques/understanding-impedance

Very nice article - super clear, and easy for someone like me to grasp. Thank you!

{EDIT- quote parity - PRR}

[jjasilli]

Ditto to pdf64.  "Line Level" is a term of art which suggests 2 things: Voltage & Impedance.  I.e., the input device is expected to receive a 1VAC signal  @ Lo Z (low impedance), say 20 - 600 Ohms.  Both factors matter.  The typical tube amp input stage is Hi Z, typically 220K - 1M.  Mismatched Z is not Ok.

[sluckey]

Ditto to pdf64.  "Line Level" is a term of art which suggests 2 things: Voltage & Impedance.  I.e., the input device is expected to receive a 1VAC signal  @ Lo Z (low impedance), say 20 - 600 Ohms.  Both factors matter.  The typical tube amp input stage is Hi Z, typically 220K - 1M.  Mismatched Z is not Ok.

Mismatched Z is not a factor in this case. The first device is capable of driving a load as low as 20-600Ω. Driving a 1M device is not a problem. The only concern would be if the driven device had an impedance "lower" than the capability of the first device, in which case it would load the signal down. Remember, we're talking about voltage levels here, not transferring power.

The main advantage of using low Z is that interconnecting cables are less susceptible to noise.

[shaun]

Thank you for the clarification! Much appreciated.