How does this fixed bias circuit work?

[moonbird]

Hello all --

I am looking to build a small amp for harp. After much research I have seen many references to MASCO amps. Given my amp building experience, I figured a single ended amp would be best - so of the MASCO amps I see mentioned, the ME-8 gets quite a bit of praise. Attached is the schematic I was able to find:

One of the first things that caught my eye was what appears to be a fixed bias setup for the 6L6 power tube. However as I looked at the circuit more closely I don't see a source for the -7.5V bias? Is the circuit drawn incorrectly or is there some "magic" here I do not understand? 

Would appreciate any other suggestions any may have on trying to clone this circuit and adapt it for harp. Thanks much!!

[sluckey]

Whenever you put a resistor between the PT HV center tap and ground, the current flowing thru that resistor will develop a negative voltage across it. That amp put the choke (which has dc resistance also) between the PT HV center tap and ground.

The bias filter cap is drawn reversed polarity!

[Willabe]

One of the first things that caught my eye was what appears to be a fixed bias setup for the 6L6 power tube. However as I looked at the circuit more closely I don't see a source for the -7.5V bias? Is the circuit drawn incorrectly or is there some "magic" here I do not understand?  

See how the PT CT does not go directly to ground? It is standing on the choke which is in the negtive side of the PSU. This lifts or actualy lowers the CT (and bias) below ground. Though I'm not sure if the -7.5vdc is correct.

You could always change the bias to cathode if you want, you might like it better. More compression.


                 Brad      

Sluckey beat me to it. I didn't notice the bias filter cap is backwards, that's bad.

[jjasilli]

Sluckey beat me to it. I didn't notice the bias filter cap is backwards, that's bad. Double ditto's!    Though at least we figured it out eventually.      

[Willabe]

Double ditto's!    Though at least we figured it out eventually.    

Yeah, the hard way. Pop, bang, ooze all over the inside of the chassis.    


              Brad      

[moonbird]

Well -- how cool is that? A "free" bias supply. What must the DCR of the choke be to develop the -7.5V? Is enough info shown? Also - I presume the 500k pot is the bias control right? Geeze 500K - this circuit must be able to produce a pretty high voltage to need that bigova attenuator. Is my thinking correct? How is the negative voltage calculated?

Willabe -- Your comment is very interesting!! I have a similar amp that is cathode biased -- a Stromberg Carlson AU-29 that has an octal "knockout" that I could possibly add the 6SF5 tube to for a second harp "channel". What other mods would I need to make it work aside from possibly volume/tone controls. I would need a "mix" resistor -- right? How do I figure the value for that?

Thanks very kindly guys for finding that error on the bias cap!!! Not the way I wanted my day to go. 

[tubeswell]

Interesting plate load for the 6SJ7 on this amp (R13 and R10, each from a different HT voltage)

[HotBluePlates]

R10 is the 6SJ7 plate load. R13 is feedback from the 6L6 plate to 6L6 grid.

Yes, there's some d.c. interaction; but the 1M has 0.21mA flowing through it and the 270k has 0.74mA flowing through it. There's also another 0.25mA flowing through the screen resistor. The 80v next to pin 5 of the 6SJ7 has to be an error, so go with the 1.3v next to G3 is probably more-right.

That implies 1.3mA of current through the 6SJ7, so there's some rounding in their numbers.

Regardless, my guess is feedback was desired, but using a resistor from plate to grid of the 6L6 would upset the bias, so it was returned on the other side of the coupling cap. Apparently, they wanted all frequencies to be impacted the same, hence the resistor. It was also apparently noted that including the 1M feedback resistor only marginally impacts the 6SJ7 plate voltage, because it's about 4x bigger than the plate load, but that's also big enough to not materially lower the effective plate load.

[moonbird]

Tubeswell --

Oooo great observation. What tonal effect would that have?

Really like the tone of this amp. Being all pentode it has a very rich tone. I tried using the "Mic vol" pot location for a "compression" control on the second 6SJ7 but it didn't seem to do much so I took it out. It has hum pretty bad and I notice that the HT passed right by the on off switch on the tone pot so I replaced it with a carling type switch in back and and put a dual concentric Bone-Ray (ala the tube wizard) in the middle pot position.

This left the far right pot position (was ON/OFF - TONE) empty. It dawned on me that this amp is setup perfectly for a VVR in that position because the HT passes very close by and it has an HT "Y" connection build in making the premap uneffected by the VVR. I've had this VVR laying around so I decided to give it a try. I put a Kevin O'conner VVR in single ended amp I had and it sound good. This is one of the simpler 1-knob types.

Trouble is - I also have been working on putting in a grounded power cord. I am waiting on a terminal strip I need to arrive so I haven't see what and how I botched it all up yet (LOL).

[sluckey]

Also - I presume the 500k pot is the bias control right? Geeze 500K - this circuit must be able to produce a pretty high voltage to need that bigova attenuator. Is my thinking correct? How is the negative voltage calculated?

That 500K pot is a tone control. It has nothing to do with bias. There is no voltage divider path for the bias voltage so all the voltage developed across the choke will be applied to the 6L6 grid.

Assume 50mA load on the power supply. That's about 45mA for the 6L6 and the rest for the other tubes. All that 50mA must flow thru the choke. Since the choke has 7.5V dropped across it, the DCR = E/I = 7.5/.05 = 150Ω.

[moonbird]

Steve --

That makes sense for sure. I had not thought to use the entire current across the choke -- but that *IS* the only way it could be of course. Thanks for the hand holding!!

[moonbird]

Here is another version of the schematic for the AU-29 -- that I find easier to read.

If I wanted to add a second 6SF5 channel for harp to the AU-29, how many, what value, and where would I need to put the mix resistor(s).

Thanks to all.

[moonbird]

Sluckey (or anyone) --

Still puzzling this out -- in the MASCO amp schematic in the first message - What does R11 (330K) do? Why wouldn't some of the -7.5 volts be dropped across it (likewise for the 500k in the pot above it)? I am asking to make sure I find the right choke for this circuit.

thx.

[Willabe]

What does R11 (330K) do? Why wouldn't some of the -7.5 volts be dropped across it (likewise for the 500k in the pot above it)?

There is no voltage divider path for the bias voltage so all the voltage developed across the choke will be applied to the 6L6 grid.

No R to ground (it would be parallal to, ie. across the bias filter cap), the other 1/2 of a voltage divider, so no voltage loss.

The power tubes grid itself draws no current, so no voltage loss going through the 330K R and 500K pot that are in series with the bias supply.

I have a similar amp that is cathode biased -- a Stromberg Carlson AU-29 that has an octal "knockout" that I could possibly add the 6SF5 tube to for a second harp "channel".

Why a second channel? Have you tried it as is for harp?


                  Brad    

 

[moonbird]

The power tubes grid itself draws no current, so no voltage loss going through the 330K R and 500K pot that are in series with the bias supply.

Ah sure -- duh -- clever people electrical engineers!! Thanks much!!

Why a second channel? Have you tried it as is for harp?

Ahhh .. what can I say -- ahh ... bordom --- ahh ... a hole with nothing in it ... ahhh 

[Willabe]

If the amp is running fine now, just try it.

It might be fine as is or you might need to tweak it a little?


                      Brad        

[sluckey]

What does R11 (330K) do?

R11 and the tone pot are the grid return (leak) resistor for the 6L6. Since bias is really the difference of potential between the grid and cathode, there must be a dc resistance path between the grid and cathode.

If you really want another channel just duplicate the mic channel up to the volume control and 330K. Tie the right hand side of the 330Ks together.

[moonbird]

Oh sure -- I see it already has the mix resistors. Wow -- thanks for putting up with my stupid questions. I have learnt a bit today. Many thanks to all!!

[PRR]

> What does R11 (330K) do? Why wouldn't some of the -7.5 volts be dropped across it

The big question is: why seven volts?? You expect 6L6 to be biased 15V or 20V.

IMHO: that "7.5V" is what you see with a meter. Moreover, a sucky meter: note the note that says 20K/V meter. If we assume a 25V-50V range that is 500K-1Meg. This loads the 330K+500K string. With a meter, you may see half of the operating bias. Without a meter, what Brad said: the 6L6 grid does not suck.

The below-ground choke is a cute trick *when you custom-wind chokes all day*. This is a 21st century. Use a resistor for bias.

> The bias filter cap is drawn reversed polarity!

It's drawn in SAMS house style. Never trust those things.

[tubeswell]

R10 is the 6SJ7 plate load. R13 is feedback from the 6L6 plate to 6L6 grid.

Ah yes, I wasn't thinking about the 6L6 plate swing. TFT HBP

[moonbird]

big question is: why seven volts?? You expect 6L6 to be biased 15V or 20V.

Seemed a little screwy

IMHO: that "7.5V" is what you see with a meter. Moreover, a sucky meter ... 

LOL!!   

The below-ground choke is a cute trick *when you custom-wind chokes all day*. This is a 21st century. Use a resistor for bias.

More sensible to be sure - thx