Question about 'totem poll' caps, balance resistors...

[eddie25]

As I understand it, the typical resistors (I usually see 220k) across 'totem polled' filter caps are mainly to even out the voltage across each cap, and for a side benifit of bleeding.

If so, without them, how uneven would the voltages be? Would it still be ok to apply close to the doubled up cap voltage values to the top of the totem?

Hope that makes sense, I'm mostly just curious and don't have a convenient voltage source to test this right now.

[Cliff Schecht]

That's actually a pretty critical component. Without it, the first cap will at least momentarily see the full voltage, especially when it first is charging up. They tend to explode quickly if you don't equalize them. I challenge you to find a single production example of an amp that stacks electrolytics in the HV supply and doesn't put on the parallel resistors.

[Bob-I]

They tend to explode quickly if you don't equalize them.

+1

Scared the living $^*T outta me.

[Dingleberry]

I would like to add two more questions.
What about if the first supply cap is just a single cap? What are the benefits to use a bleeder resistor parallel it. Some amps do have it and some amps don't.

And what's the case with the value of the resistors?
Some amps use 100K, most 220K and some even 270K.

-T

[FYL]

What about if the first supply cap is just a single cap? What are the benefits to use a bleeder resistor parallel it. Some amps do have it and some amps don't.

A bleeder R discharges the caps thus making the amp safer.

And what's the case with the value of the resistors?
Some amps use 100K, most 220K and some even 270K.

The right R value is a compromise : too small it'll draw significant current, too large the time constant formed with the caps will be too high, leading to loooong discharge times. 220K and 270K are OK with most standard power supplies using less than 200µF total capacitance.

[JamesHealey]

The biggest benefit is the draining of the voltage when the amp is turned off. The resistor creates a time constant though so you wanna make sure the frequeny is low to stop low hum coming through.

[Dingleberry]

So the point is you should always use them. I've got a bag full of 270K/1W cc resistors, so now I know where to use them.
Tomorrow I will make all of my amps safer;)
Thanks guys for the info.

-T

[FYL]

So the point is you should always use them.

It's a good engineering practice.

But you don't need them in some amp topologies, for instance if you raise reference heater voltage by a divider connected to the B+ ladder. The divider will also act as a bleeder, you can save some board space, two eyelets/turrets/... and a resistor.

I've got a bag full of 270K/1W cc resistors, so now I know where to use them.

Many people use CCs , but I stick to flameproof resistors when it comes to power supplies.

[Structo]

Yes the metal oxide flameproofs are the resistors to use in the power supply.
And of the proper wattage.
That depends on the current draw of the amp but most tube amps shouldn't have anything less than a 2w resistor in the power supply.
For 100w amps, 3w is probably even safer.

Big manufacturers design stuff with the bare minimum of parts and quality unless it is a high end item.
So a good way to judge a device is how much heat it gives off.
Heat is energy lost or dissipated, so when they scrimp on the size of resistors, they heat up more.
An old electronics teacher told me one time pertaining to solid state devices, he said if it acts like a space heater, stay away from it, as the power supply is under designed and it could cause a fire down the road as it aged.

Carbon Comp resistors can actually catch fire and burn your house down.

While not smoke proof, the metal oxide resistors will take much more abuse in a short circuit situation than other types.

Remember that part of building and designing is to make a device as safe as possible and to reduce the fire hazard as much as possible.

That's why their are codes and accepted practices on AC wiring and power supply design and construction.

P.S.
This is mostly for the noobies as I am preaching to the choir to the old hands.

[eddie25]

I think that I've seen a few early Marshall layouts with totem polled 100uf screen caps that had no bleeders. Before they started using the 56K's there.

Is this because there was the choke in front limiting the initial current surge while the caps charged?

[Jana]

Marshall used the center tap from the HV winding to give a mid-point voltage reference for the caps. This works with a full wave bridge since the center tap of the winding is in the middle of the voltage. So, even though it appears there is not a stable mid voltage reference for the caps, there is. In fact it is very low impedance compared to series bleeder resistors.

[FYL]

Is this because there was the choke in front limiting the initial current surge while the caps charged?

Marshall places the choke between the plate and screen nodes.

OTOH, some models in the JCM and JMP lines used a derivative of the Delon voltage doubler, with caps in totem pole around the PT center tap.

This slightly more complex approach allows to use smaller-rated caps and diodes, and improves ripple as well as regulation when compared to a standard bridge. No balancing resistors are required.

[eddie25]

I know about the center tap on the mains caps for balance, I was more thinking the screen cap array, right after the choke.

http://metroamp.com/wiki/index.php/Imag ... erlead.jpg

Like in the first one. Thank you all for the replies.