Need help understanding this circuit

[taj]

Hi,

Now that I've found the schematic for this little amp I was given, I'm struggling to figure out what's going on. I've only learned the bog standard basics so far, and this seems to be a little unusual in a couple spots.

Could someone knowledgeable walk me through this circuit, particularly the less common features. I can't figure out how the output bias circuit works, or why there's a pot on the bypass cap, or why the choke is on the common side of the power supply. Etc. etc. I'm just curious. Perspiring minds want to know.

..Todd

[Vinylsavor]

Hi!

The bias scheme of this amp is called 'back bias'. The voltage drop due to the current through the choke creates the bias voltage. Usually this has been done through a resistor. Here the DCR of the choke has been used. Obviously the choke needs to be placed in the ground return in this case. Sometimes chokes are placed in the ground rail so they don't need high insulation ratings. This is a compromise compared to the choke in the B+.

This amp has been done like this to save cost. It is fixed bias but done without separate bias supply by this trick. I guess there is quite some residual hum. I assume the pot with the capacitor to ground in the bias rail is some kind of hukm cancelling scheme and can be adjusted for lowest hum

Best regards ... Thomas

[taj]

Thanks Vinylsavor,

I haven't turned it on yet (not enough spare time to do it properly), so I don't know if it has a hum problem. But I guess that would probably explain why the original caps have been replaced with much larger ones. 40-50 uF.

..tj

[Miles Prower]

Originally Posted by taj

Could someone knowledgeable walk me through this circuit, particularly the less common features. I can't figure out how the output bias circuit works...

Already answered, but that's a horrible way to go about it since the DCR of ripple chokes varies widely.

or why there's a pot on the bypass cap...

This looks like some sort of half fast tone compensation that reduces the high frequency response.

I've seen few schemos as positively hideous as this one. The 6SF5 just plain SUX, as it's one of the least linear triodes I've ever seen. It was intended for use as a first pre in an AM BCB or SW receiver where fidelity wasn't a concern. It can only be used for its intended purpose, or perhaps in an FX box, but not for serious audio designs.

Furthermore, it's being used with Vgk= -0.5Vdc. Since triodes tend to pull grid current even before Vgk actually goes positive, that's not enough margin. Secondly, there is no cathode bias resistor, and this "grid leak" bias scheme is highly unpredictable, as grid leakage current likewise varies widely between makes and manufacturers and age. That says "quick 'n' dirty", and a definite must avoid.

The 6L6 is not something I'd want to use in a SE design, operating as a pentode, with limited NFB. The 6L6 tends to make lots of nasty higher order harmonics, and definitely needs the extra help, and could stand some gNFB. I also question the operating point, as Vgk= -7.5Vdc looks to be awfully low for this type.

I give this design:

out of five.

I haven't turned it on yet (not enough spare time to do it properly)

Don't.

But I guess that would probably explain why the original caps have been replaced with much larger ones. 40-50 uF.

Replacing the original capacitors with 50uF units is a really bad idea. Hollow state diodes weren't made to source the current demand that bigger filter capacitors will require. The last hollow state design I did used a 5U4GB (mucj beefier large signal diode) and I couldn't go much higher than 34uF (two 68uF/300Vdc units in series) without busting the Isurge spec (1.0A/plate).

You'll burn up the 5Y3 unnecessarily fast with excessive filter capacitance.

[dcgillespie]

The choke in the negative lead filters just as effectively as if it were in the positive side of the power supply. The pot with the cap to ground is the tone control. Moving the slider up (on the schematic) produces increasing treble cut, while moving it down ultimately eliminates it.

This bias scheme is effectively cathode bias, albeit supplied by a choke instead of a resistor. It works well in any class A design where current is constant, as it would be here. In class AB designs the circuit must be treated as a cathode biased amplifier, as increasing current through the choke produces increasing bias, as surely as it does in a normal cathode bias design, where bias is provided by a resistor.

Dave

[taj]

Since these Masco amplifiers are typically treasured by harp (blues harmonica) players, and sell for fairly silly $ on eBay, I think I'll just make sure it works as it normally would, unload it, and spend the money on parts.

Or I could use the chassis/iron to make a little guitar amp with more normal circuits.

Thanks for all the info folks!

..Todd

[Miles Prower]

^
|
|

If there's actually a demand for these thingies, then definitely restore it with the original specified filter capacitors, and E-bay it for parts.

[Tom Bavis]

The -7.5V on the grid is bogus, as it's supposed to be measured with a 20K Ohs/volt meter, presumably on a 10V or 15V scale, where it will present a load of 200K or 300K. But the bias is applied through 830K of resistance, and the meter will load that down a LOT. DON'T measure the grid bias with such a meter (Simpson 260) and leave the meter connected... output tube will COOK!

It appears that pin 1 of the 5Y3 was used as a tie point for the choke, 400 Ohms to ground, which is a reasonable value for a 6L6 cathode resistor. I'd expect -22V or so. Nonlinearity of the 6SF5 is negligible at the low level of the first microphone stage, which uses "grid leak" bias.

This could make a nice little guitar amp, should have adequate gain.

[taj]

Harp players prefer the more powerful models (push-pull 18W) models, but I suspect this 8W model will still attract attention. (8W can still get plenty loud when distortion isn't a concern!)

..Todd

[taj]

Originally Posted by Miles Prower

Replacing the original capacitors with 50uF units is a really bad idea. Hollow state diodes weren't made to source the current demand that bigger filter capacitors will require. The last hollow state design I did used a 5U4GB (mucj beefier large signal diode) and I couldn't go much higher than 34uF (two 68uF/300Vdc units in series) without busting the Isurge spec (1.0A/plate).

You'll burn up the 5Y3 unnecessarily fast with excessive filter capacitance.

Yup, that was my first thought when I opened it up and saw the big caps. But they've been in there for many years, old cardboard covered axial ones.

..Todd