Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: ampcabinets on September 11, 2010, 07:22:35 pm
-
x
-
You can tap off the high voltage winding with a .047uF 1000V capacitor like Marshall did in a lot of amps.
Resistor divider works too if you are not using a bridge rectifier.
-
I just ran into this, via a different road. A voltage doubler will work fine for the bias supply. (Then you'll be having to get rid of extra volts!)
IF you pull the feed to your bias rectifier from one of the HV legs which uses (I presume) EITHER a 4-diode bridge rectifier for the main B+ OR a voltage doubler itself, you'll almost certainly need to place a cap before the bias rectifier diode, kind of a fat one, like a .2, with a voltage rating adequate to handle your full B+. Referring to a Bogen CHB100:
http://schematicheaven.com/hifiamps/bogen_chb100.pdf
Where a doubler is used for the primary B+. Note the .22 going north-south into the. bias supply.
There's something about the condition where you have 2 power supplies but the higher current one does not have a transformer center tap. The two supplies kind of lose their ground reference. Before I put that cap in (voltage doubler main rectifier, bias taken from one HV leg) I couldn't get more than about 3 volts in the bias supply.
This phenom is used *to advantage* in high-voltage tube-based bench supplies, such as Heathkit IP-32, hp 711A, hp 712B. These are 400-500 volt, 100-200 ma supplies. There is the main, primary supply which is the 400 or 500 volt deal. That supply most often starts with a diode bridge or doubler. Then, there is a second supply, primarily used to supply bias, typically 0 to -100 or -150 volts. The two supplies are on completely separate transformer windings, but the overall schema takes advantage of the fact that there is no real "ground" reference until deeper into the circuit of either supply. Sometimes there are control tubes (like 12AX7s) in these supplies that operate with zero or single-digit volts on their plates and negative -150 volts on their cathodes! It's all the same to the poor electrons that have to do the work.
-
From the mhuss Hiwatt site---- http://hiwatt.org/Schematics/BiasCkts2.gif (http://hiwatt.org/Schematics/BiasCkts2.gif)
-
My soundcity has 550v on the plates and I had to make a voltage doubler bias circuit.Your way will work fine,but you only need one diode to make it a negative supply.
Did you change your PT or drop the voltage?
-
That will work and give approx -50vdc unloaded. The 1N4007 diode is not needed.
-
You can tap off the high voltage winding with a .047uF 1000V capacitor like Marshall did in a lot of amps.
Resistor divider works too if you are not using a bridge rectifier.
I do this all the time. I had a 100 volt bias tap but for some reason the plate voltage was way off even after getting the bias in the range i needed. I went back to the good ole high voltage tap. Under load my 100 volt bias tap was reading 60 volts so i scraped that and used the HV tap instead
-
> is there something inherently wrong with this?
The "1.4M" looks odd. That's 1,400K. The 50K trimmer, 1,400K-1,450K total, is not enough variation to be inneresting. In a plan like this, the fixed stopper resistor (prevents you going to zero bias and melting tubes) would usuallly be roughly the same size as the trimmer. If I didn't feel like thinking hard, I'd drop 47K in there.
The unloaded voltage is, as sluckey says, ~-50V at the first cap. Assuming 15K, grids, 50Ktrim+47K, the maximum (coldest) bias is near -43V. Depending on G2 voltage, that may be enough.
> would be excruciatingly loud
120 Watts is not THAT much louder than 45 Watts.
But measure your voltages. Measure power tube currents, at idle and also at FULL SCREAM. The full-power current will jump around a lot, because guitar is transient and because the transients confuse a DMM. But in a HIGH power amp, the cathode current should go WAY up when roaring. If not, you don't have enough drive or enough load. (OT impedance should probably be under 2K.)