Vinyl Savor's Octal 6GL7 preamp?

[deicide67]

I'm going to build a preamp similar to the one Thomas built here.

VinylSavor: The Octal Line Preamplifier, Part 1 : Signal Section

I'm going to use 6GF7A tubes instead which is very similar. I have a power transformer from a Fisher 400 I will never use. After the doubler, I will have just over 400V at C1 and probably a bit higher than his specified 350V max B+.

I'm going to swap the plate resistor for 6.8K which should still slow me to bias at or around his suggested 25mA idle current. I'm also considering a CCS over the fixed resistor on the cathode using LM317.

Is there anyone who has tried this circuit? If so, what are the plate and cathode voltages I should aim for?

Thank you!

Blair

 

[Vinylsavor]

Hi Blair,

the 6GF7A should work nicely in this application too.

400V will work fine. I had chosen 350V or less so I can use 350V rated caps.

I would suggest to build the circuit with cathode resistor first. Try the same 1k as in my circuit and see at which op point the tube settles. Once this works you can still experiment with a CCS.

Best regards

Thomas

 

[deicide67]

Hi Thomas,

Why did you ground the components on triode 1?

Thanks!

Blair

 

[Vinylsavor]

Hi!

Because they are unused and I would not want them to float at an undefined voltage

Thomas

 

[deicide67]

I know this is old, but I'm building this finally. How many volts should I see on the cathode?

Ive got a lower B+ of about 225V and around 35V on the cathode.

 

[Vinylsavor]

Something must be wrong. check your wiring. Did you build it as drawn? Than you would have 35mA. Most of the voltage would drop across the plate resistor. There must be a short somewhere or the tube doesn't get negative bias

Thomas

 

[deicide67]

I used 6.8K plate load resistors and a LM317 for a CCS.

I just looked, and the B+ is 225V, plate voltage is 100V, and cathodes are at roughly 60V.

That cathode voltage just looked awfully high.

Thanks!

 

[Vinylsavor]

Hi!

Best to stick to the circuit as designed unless you know exactly what you are doing. Once the original design is working as it should you can still experiment with changes

Thomas

 

[GoatGuy]

I agree with Vinylsavor… lose the CCS, go back to resistors. There is no point in optimizing a circuit "from the seat of one's pants" before the thing works as designed, first.

NOTE that the 1 kΩ cathode resistor if you are using the second, low μ section should have about a 25 volt voltage lift off ground. Quantitatively, this is E=IR … I=E/R = 25/1 kΩ = 25 ma. Which is the recommended operating point for the 2nd triode.

And the plate is supposed to be happy with 200 to 300 volts at the plate. Since you say you have a 400 V B+ supply, then the 4.7 kΩ resistor ought to have 117 V drop, nominally dissipating 2.9 watts. I'm sure that's why a 20 watt resistor was specified. Headroom.

The 2nd triode section (low μ) will have an effective plate voltage of 257 nominally.

All nicely within spec.

The 1 kΩ cathode resistor in parallel with 100 μF gives a 1.6 Hz roll-off … clearly deep in the subsonic. A capacitor ½ to ¼ the size could be used.

GoatGuy

 

[deicide67]

Thanks guys! I ended up with a different transformer, so I only have around 225V B+. I could go across the winding and cap off the center tap for 450V B+.

That's why I used a CCS. To make sure I pulled 25mA. The issue is that I've never seen a cathode voltage sit up at 60V.

I will rework the supply to get the B+ up. I was hoping not to have to use 500V caps.

 

[deicide67]

I ended up redoing the power supply across the whole winding for roughly 400V after some tweaking. The issue is that I'm still seeing crazy cathode voltage. About 60V on the cathode using the called for 1K resistor and 4.7K anode resistor.

I'm using a 6GF7A as the tube.

Pin 6 - anode
Pin 2 - grid
Pin 3 - cathode

I've built dozens of preamps, and I am stumped on this one. I swapped tubes out also in case I cooked the ones with the CCS.

No such luck.

Any ideas?

Thanks!

 

[deicide67]

I ended up redoing the power supply across the whole winding for roughly 400V after some tweaking. The issue is that I'm still seeing crazy cathode voltage. About 60V on the cathode using the called for 1K resistor and 4.7K anode resistor.

I'm using a 6GF7A as the tube.

Pin 6 - anode
Pin 2 - grid
Pin 3 - cathode

I've built dozens of preamps, and I am stumped on this one. I swapped tubes out also in case I cooked the ones with the CCS.

No such luck.

Any ideas?

Thanks!

 

[Vinylsavor]

Hi!

The circuit is very easy and should not give such problems. As mentioned if you change things you should be knowing what you are doing. I have not checked the data of the 6GF7A, but if you change from the 6GL7 you should calculate the appropriate op point and cathode resistor for it.

Also giving some more data would make it easier to help. Like what is the resulting B+ voltage, what is the plate voltage. If you have 60V across the 1k that would mean 60mA which in turn would result in almost 300V across the 4,7k and no voltage across the tube. This indicates a short somewhere

Thomas

 

[GoatGuy]

I ended up redoing the power supply across the whole winding for roughly 400V after some tweaking. The issue is that I'm still seeing crazy cathode voltage. About 60V on the cathode using the called for 1°K resistor and 4.7°K anode resistor.

I'm using a 6GF₇A as the tube.

VinylSaver has “called it”:

For cathode resistor
E = I • R
60 volts = I × 1,000 Ω
I = 60 ÷ 1000 = 60 ma.

For anode resistor
E = I • R
E = 60 ma × 4.7 kΩ
E = 282 volts

E_total = E_k + E_a + E_valve
350 volt = 60 + 282 + x
350 - 60 - 282 = x
x = 8 v???

Which is why he said, “you must have a short somewhere”. I agree. Find the short.

GoatGuy

 

[deicide67]

Thanks guys!

There is no short. I looked it over this AM, pulled a single tube from both sides, and then started independently changing resistors. Plus, I think the fuse would blow if there was a good short.

Here is the datasheet for the 6GF7A:

http://tubedata.milbert.com/sheets/049/6/6GF7A.pdf

It is a very similar tube to the 6GL7.

That being said, during these measurements, I had (Depending on the resistors) roughly 400-425V B+

Using a 4.7K plate resistor, I get:

1K cathode resistor = 60mA ~58.4V on the cathode
500 ohm = 70mA
2K cathode resistor = 50mA

Using a 6.8K Plate resistor, I get:

1K cathode resistor = 48mA ~48.2V on the cathode

It just seems odd that I have proper B+, but the cathode is so different.

The plate resistance for the 6GL7 is 780 ohms, and the 6GF7A is 750 ohms, so I would think the 4.7K plate load resistor should be about the same.

Any ideas?

I also agree that this should be a "one and done" type line stage. It is about as simple as anything I have ever built.

Thanks!

 

[deicide67]

I found this circuit for a SPUD amp using the 6GF7A:

http://zebra01.web.fc2.com/6GF7A/6GF7A_Circuit.jpg

The output transformer is close to the 4.7K plate load resistor at 5K.

They use a 5K cathode resistor, and the circuit is biased at 35mA.

I put the plate resistor value of 4.7K back in place, and used a 6.8K cathode resistor. It is what I had handy.

I got a B+ of ~440V, a Plate voltage of 275V, and a cathode voltage of 239V.

By calculation for cathode current, that is 275 / 6800 = .03514 or 35mA.

Does that look about right?

Are the two tubes that different, or am I still doing something wrong on this one?

Thanks!

 

[deicide67]

I got it! Still not sure what the hell it was, but I'm good on voltages now. I just need to get the ripple out of the PS.

Thank you for all the responses!

 

[oloppolo]

Hi. I have buit with 6EM7. I have subsonic movement in woofer. I'm looking for my mistake. Useful tips?
Thanks
Davide

 

[oloppolo]

TODAY I FOUND THE SOLUTION: CHINESE DIODES IN5408 WERE DEFECTIVE

 

[gideon1990]

Glad you fixed it! do you have some pictures for us to enjoy?