hoffman ab763 board with bridge rectifier

[embotone]

Can I use the fender bias circuit on the Hoffman board, fed off one side of the HV winding, through a blocking cap and an appropriate bias range resistor and/or by adjusting the bleed resistor off of the bias pot to get a bias voltage?  Is it just a blocking cap that lets me use one leg of a non-center tapped winding with a bridge rectifier on it that allows me to get a separate rectified voltage off of it (through a diode)?

[HotBluePlates]

You could use a cap, or not.

Normally, a large-value resistor is used to drop the excess voltage. A typical value might be ~220kΩ.

You could look at that resistor as half a voltage divider. The only problem is the other half of the divider has non-linear components, including a diode and charging pulses for the bias cap, so it's harder than normal to calculate the needed value of resistor to drop voltage.

If you're having to experiment with an unknown/untested circuit, maybe you should use a 1MΩ linear pot in place of the 220kΩ resistor, and wire the pot as a rheostat. Adjust til you get proper bias circuit operation.

You could use a cap instead of a resistor to drop the voltage. It's not really blocking anything, because it takes its input voltage from the PT secondary. Therefore it is on the opposite side of the rectifier from the B+ voltage, and there should be no d.c. present.

Instead, what it does is form a capacitive voltage divider with the bias cap. A small-value cap looks like a bigger impedance (at a given frequency) than a larger-value cap. So using two caps as a voltage divider, the a.c. is divided by the impedance of the caps, while the d.c. present at input, "ground" and the junction of the caps are all isolated form each other.

The value of the capacitive divider is that it divides a.c. while leaving d.c. unaffected (or blocked). You'll also occasionally see capacitive divider in parallel with resistive dividers in voltmeters/o'scopes, to ensure parasitic components of the resistive divider don't cause differing results at different frequencies.

A far less important benefit in your case is the physical heat dissipated by the caps will be very much less than the heat dissipated by a resistive divider, because the voltage and current are not in-phase.

So in your case, you could use either. It will probably be easier to dial in the bias circuit operation with a dropping resistor rather than a cap-divider (you'd have to have a lot of cap values on hand to substitute to find the best value).

[stratele52]

embotone,

If I understand your question I say YES and Marshall use this circuit often . Look here ;

www.freeinfosociety.com/electronics/schemview.php?id=470

[kagliostro]

HBP

Premise

I've understand this thing partially and my brain is still computing

----

you'd have to have a lot of cap values on hand to substitute to find the best value

what about a pot connected as a rheostat in series with the cap-divider ?

rotating the pot isn't similar to go on trying different cap values ?

Stratele52

as told I've understand this thing only partially but to me didn't seems the cap-divider is as the circuit used by Marshall

K

[stratele52]

Kagliostro

I though you only want to know how to use hi voltage secondary for bias circuit .I understand now you want to know how capacitor divider work.

[embotone]

The issue is taking the voltage off of one side of a winding with no center tap.  Hooking it up  off one leg of the secondary, as one would do in the case of a full wave rectifier with center tap, does not give any dc voltage of off the diode of thte bias circuit when there is a bridge rectifier and no center tap.  I have done a search and it has been discussed here before but I am still a little confused.  Perhaps I don't understand what you guys are telling me.

[kagliostro]

Embotone

sorry for the invasion of your tread

I was asking to HBP if instead to have to try a lot of capacitor as to find the exact value of the cap-divider, the addition of a rheostat (or two) in series with the caps and the turn of the same will act as a change in cap value and so in different output bias voltage (if I've understand the cap-divider affair)

Stratele52

I though you only want to know how to use hi voltage secondary for bias circuit

the original question was posted by Embotone, no by me

K

[stratele52]

You are right  Kagliostro , sorry

[HotBluePlates]

The issue is taking the voltage off of one side of a winding with no center tap. ...

Ah, yes! You are right, and I overlooked this aspect.

How about something like this (page 2).

In this circuit, 2 caps are used to feed both secondary ends' voltage to a 2nd bridge. You could then divide-down this voltage (which would start as the same as B+) to the level you need for your bias.

The catch is you need to pay attention to the orientation of the bridge. The + output gets grounded, while the - goes to the circuit.

Yes, I think you can use a resistor to help waste off voltage.

I thought I'd seen a circuit with a single cap to connect to the bias circuit in an amp using a bridge rectifier, but I can't seem to find an example of it now. Maybe I'd just seen it in an O'Connor book...

[embotone]

Here is what I found in TUT 1.  I already built my board and have a transformer suitable for a bridge rectifier on my shelf. I knew that taking a bias supply from a winding with no center tap and a bridge rectifier like this was a little different, but I didn't really think about it until I was in there wiring up the bias supply.  That's when I had that "oh crap, I forgot about that" moment.  So I really would like to stick with whats on the board as much as possible.  This amp is for someone else and I would like for it not to look like a Frankenstein/prototype build.  There are no values in O'connor's drawing but I think a .1uf 1kv cap will work as a blocking cap.  I was just hoping I could hook it up directly to the fender bias circuit with out changing the layout much.
Perhaps I should just getr in there and mess around a bit, but I hate having to muck things up too much.

[HotBluePlates]

Yes, that's the diagram I half-remembered.

A 0.1uF cap seems reasonable, and I think you're safely overrating the cap if you use a 1kV cap.

Thinking about it more, I can show you how the bridge circuit is always providing a ground for the secondary, but it is immaterial; follow the diagram you have now.

If you use a pot to experiemntally derive the values for the voltage divider, think about the possibility of scaling up the values later. You could probably figure a good ratio between the resistors of the divider using a 250k-500k pot, but if you scale up the values to larger resistances (and same ratio), less current will flow through the divider. That allows lower-wattage resistors.

[embotone]

Well, I tried placing a .1uf cap between the high voltage ac and the bias range resistor on the Hoffman board and I could not get a DC voltage after the diode.  Guess I'll give the O'connor circuit a shot, although I don't see why that would make any difference in just obtaining some dc

[kagliostro]

As I can see I've understand noting previously about cap-divider

K

[sluckey]

This is a proven circuit...

[kagliostro]

This is the method used on AC50



K

[embotone]

Thanks, I'll go with that.

[embotone]

I've installed the jcm900 bias circuit recommended by Sluckey.  Right now I get about -18vdc, I have not experimented with any other resistor values yet, but hopefully I can get the voltage up.  However, with the bias circuit hooked up, my B+ voltages rises.  Right now I have a 22uf cap off the end of the bridge rectifier to ground, with no load  added.  Without the bias circuit connected, I get 450vdc.  With the bias circuit connected, I am getting a B+ of 613vdc.  Can anyone explain to me what is up with that?  Is this an issue of not having any load off the end of the bridge rectifier?

[stratele52]

_18 is not enough . All amp bias working close to -30

Dont put output tubes in amp before.

[stratele52]

What do you mean by bias circuit connected ?

Input from transformer secondary only ? Or also connected to output tube's grids ? 

If all and complete Bias connected , you need a bigger vallue than the 50K resistor to ground until you have more than -40 volts with your bias pot.

No output tubes in the amp before please.

[embotone]

I am in the process of building this amp.  I have wired up everything on the tube socket side of the board.  I have not wired anything up on the pot side of the board.  I came to the end of the board where I was connecting up the bias circuit and realized I am dealing with a bridge rectifier and a secondary winding on the power transformer with no center tap (hit forehead with hand and say "Doh!").  I am just trying to get that sorted out before I move on.  There are no tubes installed. I have tacked on a 22uf cap between the bridge rectifier and ground to check my unloaded B+.  Everything around the power transformer is kinda temporized right now.

[stratele52]

Nice work embotome,

Did you put an .047 electroytic cap or non polarized cap  ( just before the diode ) on your bias circuit ?

You write you have more B+ voltage ( 600 v) with bias circuit connected than 400 B+  non connected ??


Bias circuit must not change B+

We talk about amp main B+

[embotone]

I used a non polarized  film cap between the high voltage and the diode.  My B+ is about 450VDC coming off the bridge rectifier without the blocking cap connected to the transformer, after connecting it between one leg of the secondary winding of the power transformer and the diode of my bias circuit, the B+ coming off the bridge rectifier rises to 613VDC.

[stratele52]

There are something wrong , B+ should stay at 450 volts. It look like you build a voltage booster.

[embotone]

Therein lies the problem, sir.

[stratele52]

Therein lies the problem, sir.

Sorry I don't understand what you write , I'm french speaking Canadian. Try in other word maybe ?

[embotone]

What I meant to say is: Yes, it looks like I have a voltage booster, and that is the problem.

[stratele52]

Thank's embotone now I understand.

You have to found why this circuit do that .

II I was you  I would replace the 0.047 cap by a 200 k or 220 K resistor and see what happend.

[stratele52]

You have probably a wiring error. Circuit with non polarized cap like you use must do the job.

[embotone]

I have removed the marshall bias circuit and substituted in the basics.  A blocking cap off one leg of the HV secondary.  Connected to a diode, with a electrolytic cap to ground on the other side of the diode.  With just this, I get my 450VDC B+, but no bias voltage.  When I add a 220k resistor to ground between the blocking cap and the diode, my B+ goes way up and I get a low and dropping bias voltage.  If I move the 220k resistor after the diode, I still get a raised B+, although not as high but rising, and a lower B-, which is dropping.  I didn't leave things on long enough to see where the voltages would settle.
Just to rule anything out, I tried another bridge rectifier and even swapped in another identical transformer.  neither of these changes improved anything.

[kagliostro]

Can you draw (at last using a pencil) the circuit you build and post it ?

Often seeing things instead to imagine it help a lot to understand what is happening

K

[stratele52]

Can you draw (at last using a pencil) the circuit you build and post it ?

Often seeing things instead to imagine it help a lot to understand what is happening

K

+1
But I think embotone will draw the  circuit ( show before and think he built ) without the wiring error. And  there are some error IMO

[embotone]

please excuse the drawing

[embotone]

This is not the bias circuit I plan on using, it is just the simplest thing I could put in there that would help me figure out whats going on.  I get the same results (high B+) when connecting the marshall jcm900 bias circuit.

[Willabe]

sorry, I posted the same image twice

If you click on the modify button for your post when it comes on the screen look at the bottom left and you'll see a little box with a mark in it for each drawing you posted. Unchech the 1 you don't want and repost.


                Brad      

[Willabe]

Do you still have a 22uF cap to ground hanging off bridges B+?


                Brad      

[sluckey]

The JCM900 bias circuit does work fine, but it's finicky. Build it EXACTLY as shown. The pot is 25K (I believe). The output of the bias circuit will connect directly to the junction of the two 220K grid resistors for the output tubes. Remove all of the original AB763 bias circuit.

BTW, Marshall specifies a Class X cap for the .047uF.

PS... You may need a 220K/2W resistor parallel to the filter cap on your bridge.

[embotone]

Willabe, yes, the 22uf cap is conncted.
Sluckey, when I had the marshall circuit connected, it was per the schematic, except I used the 10k pot I had on the board from the AB763 arrangement.
I have a general idea of what a class x capacitor is but not exactly.  Would a ceramic disk cap work here?  Can you fill me in on that?  I don't live anywhere near an electronics supply house, except Radioshack.  If I have to mail order things to make this work, I was thinking about getting one of these www.radioshack.com/product/index.jsp?productId=2102494&filterName=Type&filterValue=Transformers
and connecting the 12.6 side to my 5v winding on the transformer (yes, it has one, even though there is no center tap on the high voltage winding).  Then I could connect the other side of that tranny to a bridge rectifier and I think I should be able to come up with about 67vdc for my bias needs, if I remember my calculations from last night.

[kagliostro]

about x capacitors

EMI General Standard

K

[stratele52]

IMO you put wrong capacitor in the middle of diode brige . It look to be a polarized cap ( filter cap ) remove it and try with no cap there.

Anyway you don't need a cap there.

Be sure the other cap .047 is not a filter cap to.

[kagliostro]

I remembered something I've read some time ago in Merlin's Power Supplies book

Capacitor Coupled Bias Supply

"The bias supply relies on the bridge rectifier and the high voltage load impedance to work"

"If no load is attached to the high voltage output then no negative voltage will be developed; the circuit will instead act as a charge pump and will increase the positive voltage to as much as twice the peak AC voltage, with consequent danger to reservoir/smooting capacitors. To prevent this happening C3 should be at least ten times larger than C1 and R3 should be not more than ten times R1"


Click to enlarge


Click to enlarge

K

Merlin's books are full of interesting things and my council is to have both
(the problem is that when I put something on my brain, not always it sits there)

[stratele52]

Kagliostro ,

Yes there are voltage doubler and trippler . I think that's happened with the bias circuit if it embotone use  an electrolytic capacitor ( wrong named as filter cap ).

Filter capacitor is only one use of electrolytic capacitor as you probably know

[embotone]

I added the before mentioned transformer and bridge rectifier off the 5V winding as my source of bias voltage and it works well.  I attached the output of the bridge rectifier to the AB763 bias adjustment circuit, and with a 18K resistor off the adjustment pot I have a range of -27vdc to -45vdc.  Done deal.  Deriving the bias voltage off the high voltage winding with a bridge rectifier is an interesting thing.  I feel like I took the easy way out instead of determining my problem, but I can now move on.  From Kagliostro's attachment, I see I was missing R3.  Thanks to everyone for their help.

[stratele52]

Good news embotone, everything work with this new transformer.

I'm curious to know what was the capacitor  ( 22 mfd ) you use in the middle of the bridge rectifier on the bias circuit you do before ; Electrolytic or signal ?

I'm pretty sure this was an electrolytic and this was your problem.

[embotone]

electrolytic, but isn't that just acting as a filter capacitor?  You can't get the correct dc voltage without a cap like that on the dc coming off of the rectifier, can you?

[stratele52]

electrolytic, but isn't that just acting as a filter capacitor?  You can't get the correct dc voltage without a cap like that on the dc coming off of the rectifier, can you?

That is your problem of over voltage , you built a voltage booster . Remove your new transformer an built the bias circuit as before WITHOUT  this 22 mfd.

You have a chance to learn how to built the corect way. Bias circuit like that is basic and we use it very often. have good building habit now.

[embotone]

1.  That 22uf filter cap is the first filter node of the power rail.  Why should it not be there?
2.  Why is my method of obtaining a bias voltage inferior to taking it off the HV winding?  The transformer only cost $4.99.  It is small and unobtrusive tucked behind the power transformer.  There is more than one way to skin a cat, I've got my cat skin. If I ever build an amp with a bridge rectifier and no 5 volt winding, I'll will build it the other way.
But, as I said before, I really do appreciate your help and advice on this issue.

[Willabe]

stratele that 22uF electrolytic cap that's hanging off the bridges B+ is the reservoir filter cap. It just looks funny because of the way it's shown in the drawing.


              Brad      

[stratele52]

embotone ,
I understand that your are happy with your new small transformer for bias . Yes you can keep it if you don't want to fix the bias circuit problem you had before and you like to have an amp working as soon as possible . I understand you are not happy to re- built the first bias cicuit that give you too much B+ voltage just to see it will work without the 22 mfd cap.
________________________________________________________________
embotone and Willabe,

Where do you read you must have a 22 cap  reservoir filter cap for this bias circuit ? IT IS A WRONG INFORMATION

YES if you want a voltage booster , NO if you want the bias circuit you need.

[Willabe]

stratele,

I think we're talking about 2 different things? I was talking about the 22uF cap across the B+ bridge. I must have misunderstood you?

There's 2, 22uF caps in his drawing, 1 is across the B+ bridge and the 2'nd is in the bias circuit to ground.

Which 1 are you talking about?    


                  Brad      

  

[kagliostro]

The cap that is across the bridge image is simply connected between + and ground

look to the attached image, aren't the same circuit ?

didn't you recognize a filter (or reservoir) cap in the cap connected between B+ and ground ?

The cap draw inside the bridge is connected between + and ground, not between the AC branch of the transformer

K