Question about Hoffman 5f6a bias circuit

[schoolie]

Hi, I'm a beginner and have a silly question about the bias circuit used in the Hoffman 5f6a:  What is the function of of the 1KOhm resistor before the diode?  What I'm wondering is why not just increase the value of the 15KOhm resistor? Does it serve a function other than as as part of the voltage divider?  I'm building my own 5f6a and don't know whether it is necessary.  If I just have the 15K resistor/50K pot and 46K resistor, would it provide a suitable bias voltage range?

Thanks!

[stratele52]

The 1 k resistor is needed to lower the AC voltage going to the bais rectifier circuit.  If you go higher with this resistor , you won't have enough bias voltage. And if you remove this resistor you'll have too much bias voltage and probably over bias filter capacitor spec.

The 15 k have not the same use. This bias circuit is well made , built it as it is.s.

[EL34]

The generic board installation instructions on the library page decribe how to adjust the bias range voltage

[Ed_Chambley]

The 1k is the bias dropping resistor.  It is the one you will experiment with to get your bias in range by checking pin 5 voltages.  You will need to get around -45 on pin 5 of your output tube socket.   I just built one and simply put clips on the turrets so I could check different resistors in order to get my bias centered to the adjustment pot.  I mounted a 25k pot to the back of the amp with bias probes and just ran wires to the board.  I ended up with a 6k8 resistor in this position.

The 15k resistor, if I am not mistaken, is to limit how low the bias can go.  Kind of like the 15k resistor on the pot of a AB 763 circuit.  If the 15k was not present and you had a pot failure, you would have a meltdown of your power tubes, rectifier tube and who knows what else, expectantly if your fuse is a roll of aluminum foil.

I ended up with the bias probes, tubes installed with the 25k pot having a bias adjustment from -30mv to -46mv which gives me a nice range of adjustment so tube swaps and rebias are now a 3 minute project.

As I mentioned, I just completed one of these builds and I couldn't ask for it to be better.  It really is a nice layout.

[schoolie]

Thank you for the replies.  I have no doubt that any change to the bias circuit is for a good reason.  I was just trying to understand why the range resistor is before the diode.  Since the input is 50VAC, would it need to be reduced?  I think I understand that the 15K resistor isolates the two filtration stages, so can't be changed.   Could the 47K be changed to change the sweep?  Sorry, if these questions seem silly, I just want to understand this.

 I'll use this version, and thanks again.

[schoolie]

Thanks for the tips, Ed!  I agree.  The Hoffman 5f6a layout makes a lot more sense to me.    I'm stuck with the original layout for now, but might replace it when I can make turret boards a little more quickly.

As I mentioned, I just completed one of these builds and I couldn't ask for it to be better.  It really is a nice layout.

[Ed_Chambley]

Thank you for the replies.  I have no doubt that any change to the bias circuit is for a good reason.  I was just trying to understand why the range resistor is before the diode.  Since the input is 50VAC, would it need to be reduced?  I think I understand that the 15K resistor isolates the two filtration stages, so can't be changed.   Could the 47K be changed to change the sweep?  Sorry, if these questions seem silly, I just want to understand this.

 I'll use this version, and thanks again.

I would think the diode would not need reduced voltage.  I just know the bias range resistor being where it is is very convenient.  I don't know if this is why Doug did it this way, but it really simplifies clipping in different resistor values without the need to solder and remove until the correct one is found.  Sure changing the 47k would change the sweep, but the way it is laid out you can put everything in place except the 1k resistor.  Put your meter on pin 5, center your pot and clip in resistors until you find the correct one.  Solder it in and you are done.  It makes it so simple and removes a lot of guessing.
I may have this all wrong, but it worked so easily for me.

[schoolie]

OK, that makes sense.  Thanks, Ed!

--Rob

[Ed_Chambley]

Rob,
Gave you some wrong info after looking at my amp.  Shouldn't go from memory.  The 47k resistor keeps the pot from grounding.  the 15K resistor connects the 2 filtering caps together.  The way it looks is if you reduced the 47k resistor you would only be reducing the the resistance to ground and not the sweep.

Sorry dude.  Someone will chime in soon and straighten my butt out.

[schoolie]

Thanks, Ed.  As I understand, the 15K resistor is on one side of the voltage divider the pot and 47K resistor to ground are on the other side, so the voltage should be something like:

Vbias ~ -50*(Rp+47000)/(Rp+47000+15000)

Rp is the pot resistance. 

Anyway, really appreciate your help!

--Rob

[PRR]

I'm not sure the 1K is essential. It may be hang-over from other designs. However it is only 12 cents and might limit damage if diode or cap goes short... may as well use it.

If turrets are a problem, twist 1K and diode to each other and install as one part. Yes, this is unsightly and bad-practice if you have a 13,000 mile tour ahead, but works fine.

[Ed_Chambley]

If I understand correctly, Doug's board has a bias range resistor.  In this case it is 1k.  I was under the assumption this value was not cast in stone, but a guide as to where to experiment with bias resistors until the proper bias is achieved.  I have attached Doug's layout showing where he has marked it as a bias range resistor.  Please correct me if I am wrong, but the way I understand it the resistor can be anything necessary to achieve correct bias and this resistor replaces the 1k.

[schoolie]

Thanks, PRR!   I should have enough turrets to add the extra resistor.

--Rob

I'm not sure the 1K is essential. It may be hang-over from other designs. However it is only 12 cents and might limit damage if diode or cap goes short... may as well use it.

If turrets are a problem, twist 1K and diode to each other and install as one part. Yes, this is unsightly and bad-practice if you have a 13,000 mile tour ahead, but works fine.

[Ed_Chambley]

I'm not sure the 1K is essential. It may be hang-over from other designs. However it is only 12 cents and might limit damage if diode or cap goes short... may as well use it.

If turrets are a problem, twist 1K and diode to each other and install as one part. Yes, this is unsightly and bad-practice if you have a 13,000 mile tour ahead, but works fine.

Obviously I do not understand this.  Why would you keep the 1k resistor if you were adding another resistor?  I thought the idea was to replace the 1k with the necessary value to bring the bias inline.  Should the resistor adjustment be made somewhere else?

[sluckey]

Obviously I do not understand this.  Why would you keep the 1k resistor if you were adding another resistor?  I thought the idea was to replace the 1k with the necessary value to bring the bias inline.  Should the resistor adjustment be made somewhere else?

PRR is not suggesting using two resistors. He's simply suggesting an alternative way to mount a single resistor if space or lack of turrets is a factor as to whether or not to use the 1K resistor.

The original 5F6A did not have a 1K resistor between the PT and the diode. It also did not have a potentiometer. Doug's 'improved' bias circuit has both. Doug's intent for the 1K resistor is to provide a convenient way to set the adjustable range of the bias pot and he clearly indicates that intent on his layout drawing. The resistor has a side benefit also. In the event the first bias cap should short, the resistor would limit current from the PT bias tap and maybe save the PT from harm.

As far as where to make the resistor adjustment to change the bias voltage, well, you can do that with any of the 4 resistors used in that bias circuit. But changing the 1K BIAS RANGE RESISTOR is usually the easiest way to get in the 'zone' if your bias tap voltage differs from Doug's suggested PT, or if you use a different type tube that requires a different range of bias voltage. Heck, you might not even have a bias tap on the PT. No problem. Just use one of the HT leads for the input to the bias circuit. All you gotta do is just use a much bigger (100K) BIAS RANGE RESISTOR.

[Ed_Chambley]

Sluckey,
Thanks for the clarification.  I thought I understood it correctly, but reading the post it seemed PRR was suggesting a need to maintain the 1k resistor.  I understand Doug's bias design and adding another e-cap for extra filtration was a cool idea.  At least I assume this is why there are 2 e-caps connected by the 15k resistor.  It seems PRR is extremely knowledgeable as I read a lot of his posts as they are very beneficial to me.  It also seems to me that reducing the voltage prior to the bias circuit has the benefit of helping to control drift as parts begin to age unless of course the drift is at the bias range resistor.

[sluckey]

I understand Doug's bias design and adding another e-cap for extra filtration was a cool idea.

Doug didn't add any additional caps to the bias circuit. Look at the original.

It also seems to me that reducing the voltage prior to the bias circuit has the benefit of helping to control drift as parts begin to age unless of course the drift is at the bias range resistor.

Say what???

[Ed_Chambley]

I understand Doug's bias design and adding another e-cap for extra filtration was a cool idea.

Doug didn't add any additional caps to the bias circuit. Look at the original.

It also seems to me that reducing the voltage prior to the bias circuit has the benefit of helping to control drift as parts begin to age unless of course the drift is at the bias range resistor.

Say what???

This is how Doug explains it:
This is the bias system that I use on the 5F6A and Plexi board kits. Notice it has one more stage of filtration than the simple Fender bias system. After the diode there is a 10uf/100v filter cap. The negative bias voltage then goes through a 15K divider resistor that isolates the next stage from the previous stage. The bias voltage then is filtered again by the second 10uf/100v cap. From there it enters the bias pot, which is wired like a variable resistor. The 47K resistor on the other leg of the bias pot ensures that the bias voltage can never short out directly to ground when the bias pot is turned to zero ohms.
Notice that one leg of the pot is not connected to anything. This is what makes it a variable resistor.

Now my concept on reducing the voltage prior to the bias circuit.  5% drift of 25volts is half as much as 5% drift of 50 volts. Bias tap is 50 volts and all parts have a variance like CC resistors usually 10%.  If less voltage is applied the percentage of voltage change has to be less.  Is this crazy or what?

[sluckey]

LOOK AT THE ORIGINAL 5F6A CIRCUIT. TWO CAPS.

If less voltage is applied the percentage of voltage change has to be less.  Is this crazy or what?

Do the math with some numbers. The percentage of change will be the same.

[Ed_Chambley]

LOOK AT THE ORIGINAL 5F6A CIRCUIT. TWO CAPS.

If less voltage is applied the percentage of voltage change has to be less.  Is this crazy or what?

Do the math with some numbers. The percentage of change will be the same.

I did look at it.   There are certainly 2, 8uf at 150.  Wonder what Doug meant?

[sluckey]

I suspect he meant AB763?

[EL34]

There are certainly 2, 8uf at 150.  Wonder what Doug meant?

meant by what?

Basman 5f6a bias circuit shows 2 x 8/150 caps
my board shows 2 x 10/100v caps

The bias range resistor is explained in my generic assembly instructions on the library page

[Ed_Chambley]

There are certainly 2, 8uf at 150.  Wonder what Doug meant?

meant by what?

Basman 5f6a bias circuit shows 2 x 8/150 caps
my board shows 2 x 10/100v caps

The bias range resistor is explained in my generic assembly instructions on the library page

I simply looked at this from the library page
This is the bias system that I use on the 5F6A and Plexi board kits. Notice it has one more stage of filtration than the simple Fender bias system. After the diode there is a 10uf/100v filter cap. The negative bias voltage then goes through a 15K divider resistor that isolates the next stage from the previous stage. The bias voltage then is filtered again by the second 10uf/100v cap. From there it enters the bias pot, which is wired like a variable resistor. The 47K resistor on the other leg of the bias pot ensures that the bias voltage can never short out directly to ground when the bias pot is turned to zero ohms.
Notice that one leg of the pot is not connected to anything. This is what makes it a variable resistor.

I assumed from this comment that your 5F6A board had additional filtration.  It was pointed out to me the original Fender schematic has 2 caps as well.  So I asked wonder what you meant by stating it has one more stage of filtration than a simple Fender bias system.
Sluckey said AB763.

This being said, I cannot find a link title "generic assembly instructions" in the library page.  What I did find, a portion is pasted above.

At this point, I think we are beating a dead horse.