Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: kagliostro on December 27, 2012, 08:25:05 am
-
I've never seen a mod like this for a 5e3 Deluxe or other cathode bias circuit, and you ?
(http://img13.imageshack.us/img13/5430/modbiastweeddeluxebias.gif)
Also if I can unserstand the reason for the 150R resistor and 1K trimmer
I'm not sure which is the action of the 500K trimmer, seems something as to do with balance, but I'm not able to understand
K
p.s.: I like to find things I don't know that make me curious, if I arrive to know what is back, to me is a great satisfaction :icon_biggrin:
-
The 500kB pot appears to be a "balance" pot for the grids....when at 50% (center of it's travel) you will have 250k in series with the 220k resistors, effective 470k, which can be tweaked one way or the other to balance the bias........this would allow one to balance the bias to "match" unmatched power tubes. Measure the cathode current across a separate 1R for each power tube cathode and adjust the balance pot until the current for each tube is the same.
The 150R in series with the 1k pot is parallel with the 250R bias resistor, allowing you to adjust the bias resistor value, ranging from ~205R (250||1150) down to ~94R (250||150)..... The 150R is to insure the bias resistor value never goes below a certain value when the 1k pot is turned to it's minimum value.
I would tweak the values of the bias resistors a bit....make the 250k larger to give a wider range of bias available. And make the 150 larger so the bias can't set so hot.
G
-
Many thanks Geezer
About this mod I've discovered it was an experimental mod, not a tested mod
---
so the action of the 500K pot has to do with the load seen by the PI and also can be seen as a way to drop a part of the signal to ground, the signal from one branch of the PI and so a balance of the output of the two tubes
may be something like to have a 100K on one plate and 82K on the other plate of the PI or something like use trimmers in place of plate PI resistors
About the use of the 150R + 1K I assumed it to be the exact contrary, I was thinking that the 150R is the resistor to be put in parallel of the standard 250R resistor and the 1k trimmer was there to exclude or progressively insert the 150R resistor in parallel, same thing, two different interpretations :l2: :l2:
I've seen a resistor to be put in parallel of the cathode resistor on Vox AC30CC2, there was a switch in series to the add on resistor to insert or exclude it from the circuit, so I've think the switch function was substituted by the trimmer
However I think a better approach is those experimented here on the forum some time ago, with the bypass resistor connected in series with a trimmer or fix resistor+switch
Many Thanks for your considerations
K
-
IMO 150 ohms and 1K pot are only to adjust bias.
500 K pot are for balance Phase inverter ouput signal to 6V6's
-
What would the wattage of the 500kB pot, the 1kB pot, and the 150 ohm resistor have to be?
-
What would the wattage of the 500kB pot, the 1kB pot, and the 150 ohm resistor have to be?
The 500 K ; same watts as 220 k ; 1/2 W
The 1K ; few watts for sure . But I did not calculate how many. A 5 watts will do the job, like the original 250 ohms 5 W
-
Remember that this isn't a tested mod
As told I discovered it is only a proposal of mod
K
-
+1 kagliostro.
The 150 ohms + 1K pot will give a very very hot bias . Too hot ? Have to take some reading.
500 K , not sure you'll hear if it is better.
-
The 500K does not change the DC bias. It may change the loading on the driver. But it doesn't make much change (220K-720K from a driver which is typically under 100K).
The 720K extreme may exceed the 6V6's max grid resistance rating, though 99% of tubes will be OK.
The cathode bias as shown can only increase 6V6 current. But a stock cathode-bias Deluxe already runs the tubes near maximum heat. If the 250r is removed, this would allow more/less than stock bias current. However that 1K is a BIG wattage. There's better ways to do this.
I think this is not fully thought-out.
-
Thanks PRR
As told I've find that mod and in a first time I was thinking to it as a tested mod, so I developed interest on understanding what was happening there
after I discovered that the mod was only a proposal for a mod, a mod idea, never build or tested
Many Thanks
K
-
I'm not sure which is the action of the 500K trimmer, seems something as to do with balance
Yes, it's purpose would be to balance the drive signals from the PI, but it may not be very effective. But if you eliminate the 220K resistors and connect the ends of the balance pot directly to the 1500Ω grid stoppers, you will be able to perfectly match the drive signals. Doing so may be worthwhile in the hifi world but is mostly wasted on a lofi 5E3.
-
I'm not sure which is the action of the 500K trimmer, seems something as to do with balance
Yes, it's purpose would be to balance the drive signals from the PI, but it may not be very effective. But if you eliminate the 220K resistors and connect the ends of the balance pot directly to the 1500Ω grid stoppers, you will be able to perfectly match the drive signals. Doing so may be worthwhile in the hifi world but is mostly wasted on a lofi 5E3.
+1
-
I think this is not fully thought-out.
I think the originator was half-remembering the Williamson biasing arrangement, while getting small but important aspects wrong, and inserting it incorrectly in parallel with the original cathode resistor.
Look at the Williamson circuit below:
I have a version of this bias circuit in my Standel copy.
You don't really need C6; use it if you want, but I didn't.
R22, R16 are 100Ω pots (I used 2w PEC pots with a locking shaft)
R23 should be your minimum desired cathode resistance. 150Ω is stock for the Williamson, I might have used slightly higher or lower. This sets the limit for how hot the output tubes can be biased.
R15, R17 are normal grid reference resistors; stock Wiliamson is 100kΩ, but you could use 120-150kΩ without issue.
R18, R19 are 100Ω resistors. They connect from the 100Ω pot in the cathode of the output tubes (wired as a rheostat), and go to either end of the 100Ω pot that sits between the grid reference resistors.
R16 has the actual ground reference for both the grid reference resistors and the cathode bias circuit components. You set the rheostat (R22) to get roughly correct idle current for both tubes, then tinker R16 to get equal current in both tubes.
Total cathode resistance seen by the tubes is the parallel resistance of R18 & R19 in series with R22, R23 and some variable resistance from R16.
I added 1Ω resistors in my amp between R23 and an output tube cathode to allow measuring idle current with a voltmeter. Where each 1Ω resistor (not pictured in the schematic below) connects to the end of R23 that previous was common to both tube cathodes, I added a meter tip jack as my "ground". Each tube then has a "hot" tip jack at its cathode. Measuring millivolts from a "hot" jack to the "ground" jack gives idle current for that tube.
My procedure:
Monitor 1 tube's idle current, and adjust R22 for rough desired current. Connect the voltmeter from "hot" to "hot" (meaning from one tube's cathode to the other) and adjust R16 for 0v. When 0v is read between the cathodes, there is equal idle current in both tubes.
I could instead measure one tube's current in the conventional manner, then the other tube's current in the conventional manner and compare, but that's slower. As one tube's idle current is being reduced by R16, the other is having its current increase. So it's much faster to measure from cathode to cathode.
I can attest from my experience with this amp that the bias adjust and balance works as advertised. But you'd have to have a really good reason for needing to tinker the bias on an otherwise self-adjusting 1-resistor method. You're adding 2 resistors (4 if you weren't gonna use 1Ω resistors) and 2 very expensive pots, plus 3 tip jacks.
-
Hi HotBluePlates
That is very interesting (every day I discover that your knowledge is larger and larger)
Many thanks for the given explanations
I've only two question for you
. Which was the principal reason for you to experiment with this bias circuit ?
. You talk about expensive pot, which are the spec for the pot you used and more, which type of pot ?
Thanks again
K
-
Hi HotBluePlates
. Which was the principal reason for you to experiment with this bias circuit ?
. You talk about expensive pot, which are the spec for the pot you used and more, which type of pot ?
K
1--Easy adjustable Bias , different tube's brand = different bias. No need to replace a resistor each time you change bias
2- 5 watts and more pots are more expensive than 1/2 w standard pot . They usually named Wire Wound pot and could cost $40 each
-
Sorry I didn't see the questions until now. My Standel build uses cathode bias, and I wanted adjustable bias and bias balance. So I used this circuit.
I used 2w PEC RV4-style locking pots, 100Ω each.