some questions about tube amp build

[kepeb]

cheers abbey,
this is a continuation of the thread I started in the brewery,
http://www.groupdiy.com/index.php?topic=46508.0
all the links are on the first page

its a physical project and is on 'the bench' so needs to come into the lab.
I have a few more questions for the experienced tube fellas.

seems the weber design leaves some to be desired in regard to the bias voltage, any more thoughts here? why would they suggest a flawed method?
https://taweber.powweb.com/store/6o100_schem.jpg

the HT fuse is wired from the Centre tap direct to ground, how does that work to stop the flow??

do i need a special pot for the presence control as this is feeding back volts from the output?

EDIT:
here are a couple of other links i have found useful when i have questions:
http://paulrubyamps.com/info.html
http://www.duncanamps.com/technical/lvbias.html

 

[hobiesound]

The bias circuit looks fine to me, although you might wanna take a look at how marshall or fender does this (which is almost the same). And i would suggest adding 1 watt 1 ohm 1% resistors from cathode to ground on the power tubes to measure the idle current.

in marshall amps the presence pot is the same as al the other pots.

the fuse to ground is fine where it is just another way of doing it.

greetings,

Thomas

 

[abbey road d enfer]

seems the weber design leaves some to be desired in regard to the bias voltage, any more thoughts here? why would they suggest a flawed method?

I have to say I don't have much consideration for Weber's "designs". His papers in Vintage Guitar Magazine show the paucity of his scientific background.
Also don't forget that very often, kit manufacturers will publish their schematics with blatant errors.

 

[gemini86]

The fuse on the centertap cits the return to ground in the entire circuit. I don't like that it still leaves the entire circuit floating at hv potential during a fault.

I once bought a zener kit from them that didn't quite do what they said it would and nobody there could explain the theory either. They do make some nice speakers though.

 

[PRR]

> the weber design leaves some to be desired in regard to the bias voltage, any more thoughts here?

What are you desiring? I'm clueless what "flaw" you are objecting to.

It's a fine design. It has the special virtue that _when_ the pot wiper loses contact, bias goes to max (safe) instead of zero (melt-down). It has the slight objection that a careless tech can turn bias to zero--- I feel it would be wise to have a fixed stopper resistor so that bias can't be turned much lower than 120% dissipation on a bogie tube (fast roast, not instant melt-down).

> the HT fuse is wired from the Centre tap direct to ground, how does that work to stop the flow??

It is a _circuit_. A break anywhere in the loop breaks the whole loop.

> it still leaves the entire circuit floating at hv potential during a fault.

How can it do that? Try it. The entire amp will be at zero volts DC. Yes, the PT ends will be swinging 300V AC; there's no way to avoid that with secondary fusing.

I'm not a fan of secondary fusing on well-built amps. But for a KIT with beginner builders it is probably a very wise feature.

 

[gemini86]

Hmmm.. my measly little brain was under the impression that everything before the open circuit would be at hv. I guess the diodes would stop conducting without a ground reference. Is that what I was missing?

 

[dai h.]

seems the weber design leaves some to be desired in regard to the bias voltage, any more thoughts here? why would they suggest a flawed method?

I have to say I don't have much consideration for Weber's "designs". His papers in Vintage Guitar Magazine show the paucity of his scientific background.
Also don't forget that very often, kit manufacturers will publish their schematics with blatant errors.

Hi. Probably true for the "Weber (G.--former meat salesman as I understood)" whom I think you are referring to, but the "Weber VST" guy ("Ted". RIP) is a different person (with the same surname) and does(did) have a technical background (forget which one, but worked for an automobile manf. and USAF previous to that as I recall).

 

[okgb]

Also spend some time researching grounding schemes for gtr amps

 

[gemini86]

I found this to be useful...

http://www.freewebs.com/valvewizard2/Grounding.html

I've employed most of it all in build before, with good results.

I've also employed poor ground schemes with "good" results as well... so there.

 

[abbey road d enfer]

> the weber design leaves some to be desired in regard to the bias voltage, any more thoughts here?

What are you desiring? I'm clueless what "flaw" you are objecting to.

It's a fine design. It has the special virtue that _when_ the pot wiper loses contact, bias goes to max (safe) instead of zero (melt-down). It has the slight objection that a careless tech can turn bias to zero--- I feel it would be wise to have a fixed stopper resistor so that bias can't be turned much lower than 120% dissipation on a bogie tube (fast roast, not instant melt-down).

The main objection is the 1k resistors, which induces excessive dissipation into the pot. I still maintain that option #3 is preferrable, with the addition of a resistor between top and wiper.

 

[abbey road d enfer]

seems the weber design leaves some to be desired in regard to the bias voltage, any more thoughts here? why would they suggest a flawed method?

I have to say I don't have much consideration for Weber's "designs". His papers in Vintage Guitar Magazine show the paucity of his scientific background.
Also don't forget that very often, kit manufacturers will publish their schematics with blatant errors.

Hi. Probably true for the "Weber (G.--former meat salesman as I understood)" whom I think you are referring to, but the "Weber VST" guy ("Ted". RIP) is a different person (with the same surname) and does(did) have a technical background (forget which one, but worked for an automobile manf. and USAF previous to that as I recall).

The schemo on the TA Weber site shows a different type of bias adjust, with 2x50k pots, which has the flaw of leaving a tube unbiased in case of wiper contact loss (as PRR mentioned).
But the schemo I consider as seriously flawed is the one from Gerald Weber, which shows 100k pots mounted as rheostats, with a 1k drop resistors, and that i demonstrated would undoubtedly exceed the dissipation limits of the pot (or require a 5-10W-rated pot).

 

[dai h.]

The schemo on the TA Weber site shows a different type of bias adjust, with 2x50k pots, which has the flaw of leaving a tube unbiased in case of wiper contact loss (as PRR mentioned).

(Maybe I'm looking at the wrong one but I'm slightly confused.) I thought this one was the TA Weber one being referred to (which shows just one 50k pot--which would just result in the negative bias voltage rising if it failed open) :
https://taweber.powweb.com/store/6o100_schem.jpg

But the schemo I consider as seriously flawed is the one from Gerald Weber, which shows 100k pots mounted as rheostats, with a 1k drop resistors, and that i demonstrated would undoubtedly exceed the dissipation limits of the pot (or require a 5-10W-rated pot).

So (if I understand correctly), this is in reference to the trim pots if which lowered far enough would dissipate too much power in the range of adjustment you'd want them to be in? Personally I don't get why there isn't a series R with the pot (for the reason mentioned previously somewhere above-- i.e. to limit the adjustment range for safety).

 

[abbey road d enfer]

The schemo on the TA Weber site shows a different type of bias adjust, with 2x50k pots, which has the flaw of leaving a tube unbiased in case of wiper contact loss (as PRR mentioned).

(Maybe I'm looking at the wrong one but I'm slightly confused.) I thought this one was the TA Weber one being referred to (which shows just one 50k pot--which would just result in the negative bias voltage rising if it failed open)

If the wiper loses contact with the track, bias is left floating and grid current will push the tubes in saturation. If it fails open between bottom and wiper, yes, bias will increase, but if it fails open between top and wiper, bias goes down to zero.

But the schemo I consider as seriously flawed is the one from Gerald Weber, which shows 100k pots mounted as rheostats, with a 1k drop resistors, and that i demonstrated would undoubtedly exceed the dissipation limits of the pot (or require a 5-10W-rated pot).

So (if I understand correctly), this is in reference to the trim pots if which lowered far enough would dissipate too much power in the range of adjustment you'd want them to be in? Personally I don't get why there isn't a series R with the pot (for the reason mentioned previously somewhere above-- i.e. to limit the adjustment range for safety).

But there is! The 1k 5W between the diode and the cap. Unfortunately it's so low, it operates the trim pots far above their limits. And this would happen in quite normal conditions, with 60V available and trimmed at 48 V, current would be 12 mA, the pot would be set at about 3% of its value, making the adjustment very clumsy. I think the schemo is deliberately (or not) faulty; I reckon this series resistor should be about 10k.

 

[kepeb]

excellent stuff, thanks for your responses

I get the HT fuse now that discussion makes sense.

I guess if marshall use standard pots for presence then that should be fine for me.

Also spend some time researching grounding schemes for gtr amps

indeed, this was really the only part i thought i was comfortable with when first approaching, i have stepped away from my previous strict star ground for the first time, and based on similar layouts, implemented a split ground across both ends of the chassis. I have read so much, often conflicting information in regard to ground wiring though. I think it will be best to see what works. I am more than prepared to make adjustments and rearrange grounds after i get it working. I dont think my heater wires are in the ideal location since finishing the wiring...There may be more to this yet as my build and controls are not best shielded either.

theres a nice checklist type page here for tube amp building too
http://www.paulrubyamps.com/info.html#FirstPowerUp


the one remaining point then is this dual bias. PRR I'm sorry to have given the wrong impression, I was not intending to insult the kit design (with few minor exceptions its pretty much the same as original orange)

you yourself emphasized the 'when' when discussing the wiper failure, leads me to believe it is unavoidable with this design. the old sc design appears to depend on the wiper completely.
presuming the following attachment is suitable and safe to use for dual bias (to achieve the -55v suggested)
what kind of value should i put in the red spaces?
maybe 10k again?

 

[abbey road d enfer]

what kind of value should i put in the red spaces?
maybe 10k again?

It depends on the rectified voltage. Put Ohm's law at work.  With R=10k and P=50k and a bias winding that delivers 50Vac, the bias voltage would be 59V at max, 54 V at mid-rotation and 34V minimum. This is for a single bias!. For double-bias, the voltage drop in the first 10k would be doubled. You have to know the specs for the bias winding if you want to do any serious calculation. The other option is to experiment, without the tubes in their sockets.

 

[dai h.]

But there is! The 1k 5W between the diode and the cap. Unfortunately it's so low, it operates the trim pots far above their limits. And this would happen in quite normal conditions, with 60V available and trimmed at 48 V, current would be 12 mA, the pot would be set at about 3% of its value, making the adjustment very clumsy. I think the schemo is deliberately (or not) faulty; I reckon this series resistor should be about 10k.

ah, I think see which schematic is being referred to (the middle one, I think). So the 1k needs to be higher. (To my novice thinking, since you just need a voltage (a tiny current) for the C-, it doesn't make sense to set the bias circuit to load the winding in a way to require any significant power).

 

[abbey road d enfer]

ah, I think see which schematic is being referred to (the middle one, I think). So the 1k needs to be higher. (To my novice thinking, since you just need a voltage (a tiny current) for the C-, it doesn't make sense to set the bias circuit to load the winding in a way to require any significant power).

You're absolutely right. These bias circuits are based on extremely low current draw - that's why they can get away with half-wave rectification. The only limitation is that the grids must see a resistance sufficiently low in order to avoid grid-current bias (the max value for the grid resistors is in the valve's specs).

 

[CJ]

a fuse on the center tap keeps you from getting across the B+ line,

most fuse holders are rated to 120/240, so an open circuit (blown fuse on the B+ line) might put 400 volts dc on the fuse holder, depending on the circuit.

law suits from customers is the #1 threat to any small bidness,

forget the bias pot and run huge cathode resistors for a much warmer sound,
those orange amps are kinda harsh, like a Laney or Hiwatt,
plus, tubes last forever since the bias is more stable.
dial in the screen volts one notch below cherry red and you are set for life as far as bias adjustment is concerned.
and you can tune the low end with the bypass cap on the push-pull cathodes,

 

[lassoharp]

forget the bias pot and run huge cathode resistors for a much warmer sound,
those orange amps are kinda harsh, like a Laney or Hiwatt,

+1.  Those 50W and higher amps seem to want to be ran balls open to overcome the brittleness and turn fat.  Works great on a big open stage.  Hell in the living room. 

 

[kepeb]

I just wanted separate adjustable bias...
they seem to have had no complaints with the single bias wired 'as is' for their kits
so, in order to have this design with dual bias... use a different design.  ??

cathode biasing would make the build I planned substantially different, particularly with a parallel cap.
this would also render the FAC control (which is specific to this build) pointless, moving a low cut to a different position.

the PTX i used is mojo 762ex (export) fender replacement. the bias winding states 53v on the diagram i found.

It may sound like i'm being awkward but i want to stay close to this orange design, its clear i'm only learning about bias  requirements and need some help or suggestions here to help me understand.

Abbey says the published design will fail, has excess pot dissipation and suggests a route for a corrected single bias (maybe easier to work out with known ptx above) but not how to dual this.

so, lets assume i'm getting exactly 53v on the bias.
from what i can find on EL34 biasing the recommended is -38v, as weber suggests 'starting' at about -55v can i also assume a more negative voltage/more resistance to ground, will be less current draw?
how do i determine the current here?

If the current (which from what i understand but is not well defined) will be about 20mA from bias to tube (each tube in push pull?), then this is 40mA to the tube grid. how does this stress the pot?

I must be understanding this wrong...
acting as a voltage divider the current would be going to the tube only and not through the remaining trim value, which will just determine a reference point for voltage?

I am searching for information on this but its not that well documented or explained as other parts of amp design.