Choke or resistor

[labb]

In the power chain. Which is better. See schematics all the time showing a choke and an equivalent value resistor. Dollar wise it is a no brainer.  Cheep resistor versus a $25+ choke.

[shooter]

it's all in the ear of the beholder, or the mind of an engineer

[stratomaster]

I always find amps without chokes to be noisier than amps with.  The noise can be overcome though--typically at the expense of the charm of the choke-less design (increased capacitance, additional filter nodes, relocating triodes to different filter nodes, etc).

[kagliostro]

Solution in the middle for expense but way better than a choke about noise reduction (or better, to arrive at this noise reduction you must invest in an heavy high quality choke), see attached sematic and layout

Franco

[HotBluePlates]

In the power chain. Which is better. ...

You need to know how current will change from idle to full-power.

The value of a choke is the filtering effect is large (many-Ω of reactance), but the resistance is small.  When there is a big current-change, the small resistance means no big change of voltage-drop across the choke.  So the power supply voltage after remains more stable.

A resistor is cheap, but if there is a big change of current from idle to full-power, there is a big voltage-sag at full power.  When the next thing after the choke/resistor is an output tube screen, then the lowered voltage chokes the power output capability of the tube.


So smaller/older amps that tend to have power sections that are Class A (or barely into Class AB) can get away with a resistor, because the current-draw at full power is similar to the idle current draw.

[Merlin]

These days we can use a resistor and a bigger cap than vintage amps had access to. A 470R resistor and 100uF cap is practically as good as a choke and 20uF in a vintage circuit. The rest is cork sniffing.

[labb]

If using a resistor does it need to be high watts. I’m thinking about what the Train Wreck builders do. They use a 25 watt resistor.

[stratomaster]

If using a resistor does it need to be high watts. I’m thinking about what the Train Wreck builders do. They use a 25 watt resistor.

P=∆V²/R

How much voltage is dropped across how much resistance? That'll determine the power. 25W is pretty high.  If you know the resistance value in a TW you can use the above equation to calculate the maximum voltage you can drop across that resistor. If this grossly exceeds the B+ in the amp then you know it's massive overkill.

Update: looked at one of the schematics. 25W at 1kΩ.  This can handle 158V across it at 158mA based on the formula above. It would never need to drop that much in a power supply.

Where it is in the schematic has the nodes for screen, PI, and 3 triodes following it. Assuming a total current draw of 25mA and doubling for good measure puts about 2.5W dissipated in that resistor. A 3-5W resistor would've been bulletproof.  You could likely even get away with a 2W, since I doubled the current estimate and the power calculation goes as the square of the current.

Going back to our equation above and shuffling it to be V=P/I, and substituting our 50mA estimate for current draw shows the 25W resistor is good for a 500V drop at the expected current. This is insane.

So yes, higher power resistors are needed.  No they don't need to be 25W in a TW. 

[Merlin]

If using a resistor does it need to be high watts. I’m thinking about what the Train Wreck builders do. They use a 25 watt resistor.

Typically 5W though it does depend on how big the amp is. 25W is just silly; 2W would be enough for a couple of 6V6s!

[labb]

My calculations same as yours. I have typically used 5 watt resistors and never had a problem. I questioned this once and was given an explanation about how the screens swing. Didn’t quite grasp what I was told but the 25 watt was justified. Different strokes for different folks. Thanks for all of the replies.

[Merlin]

I questioned this once and was given an explanation about how the screens swing. Didn’t quite grasp what I was told but the 25 watt was justified.

Screen current can certainly triple during overdrive, but that's still nowhere near needing a 25W resistor.

[GlideOn]

IMHO, it's not so much the choke/resistor as it is the mains filter cap being inadequate.

Decades ago, high voltage, high capacitance was at a premium and transformers were a bargain. Now it's the opposite.

There's no real downside to use a larger farad mains filter cap when 100uf/450v can be readily found in radial or axial config.

Decreased noise floor, increased bass response without screwing up the core tone. Keep your mains high and your pres low.

Win/win.

[Willabe]

There's no real downside to use a larger farad mains filter cap when 100uf/450v can be readily found in radial or axial config.

Yes there is if your using a tube rectifier.

[HotBluePlates]

IMHO, it's not so much the choke/resistor as it is the mains filter cap being inadequate.

They do call it Ohm's Law...

     A 4H choke has 3,000Ω of reactance, but only 108Ω of resistance.

     A 3kΩ resistor passing the rated 90mA will drop 270v.

     The 4H 90mA choke passing 90mA will drop only 9.72v.


An output tube's plate current drops when the screen volts drop.  A 270v reduction of screen current will turn a Super Reverb into a 5 watt amp.

IMHO, it's not so much the choke/resistor as it is the mains filter cap being inadequate.
...
There's no real downside to use a larger farad mains filter cap when ...

90% of the time, the choke is used between the main & 2nd filter caps, feeding the output tube screens.  The last point above is why.

Big main filter doesn't address the problem the choke is trying to solve.

[glass54]

I am fond of using the R+C method. I NEVER Exceed 470R! So I choose a 10 Watt plus AluClad resistor (mount it on the chassis) and extra uF. But I also use SS diodes or overrated Bridge rectifier 
Using 90mA as an example, worst case voltage loss with 470R is 42V, but I would go much lower and up the uF value.
I agree with HBP, you have to really think about this 
Regards
Mirek

[GlideOn]

IMHO, it's not so much the choke/resistor as it is the mains filter cap being inadequate.

They do call it Ohm's Law...

     A 4H choke has 3,000Ω of reactance, but only 108Ω of resistance.

     A 3kΩ resistor passing the rated 90mA will drop 270v.

     The 4H 90mA choke passing 90mA will drop only 9.72v.


An output tube's plate current drops when the screen volts drop.  A 270v reduction of screen current will turn a Super Reverb into a 5 watt amp.

IMHO, it's not so much the choke/resistor as it is the mains filter cap being inadequate.
...
There's no real downside to use a larger farad mains filter cap when ...

90% of the time, the choke is used between the main & 2nd filter caps, feeding the output tube screens.  The last point above is why.

Big main filter doesn't address the problem the choke is trying to solve.

Yes, obviously a choke is superior method of combating noise and voltage surges, but in the spirit of pragmatism and answer to the OPs question: If the OPs main problem is noise and hum and isn't using a tube rect, a huge cap in the first stage absolutely solves his problem and solves it well.

To the workingman customer who just wants his amp to be quieter and doesn't want to drill holes or do complex, costly modifications this is exactly what I'd do.

Presumably there is already room on the board for said cap (only slightly larger at worst) and less so for extra transformers, extra tube sockets.

If you're using a tube rectifier in a higher wattage or Class AB amp, you have other inherent problems and probably should incorporate a choke to compensate, or live with the noise warts and all.

As you yourself said, a small resistor on a simpler Class A amp is perfectly fine as the voltage/current demand difference are negligible.

At the end of the day there is no such thing as a truly perfect solution. It's about whatever solution you can live with.

[Willabe]

If you're using a tube rectifier in a higher wattage or Class AB amp, you have other inherent problems...

What 'other inherent problems'?

[pdf64]

If you're using a tube rectifier in a higher wattage or Class AB amp, you have other inherent problems...

What 'other inherent problems'?

May mean 'performance characteristics'?
 
Which can be desirable or not.

[Willabe]

If you're using a tube rectifier in a higher wattage or Class AB amp, you have other inherent problems...

What 'other inherent problems'?

May mean 'performance characteristics'?
 
Which can be desirable or not.

Nah, let him answer. 

[HotBluePlates]

At the end of the day there is no such thing as a truly perfect solution. It's about whatever solution you can live with.

A 12AX7 with a 5kΩ plate load resistor and a 10kΩ cathode resistor will pass sound.

   It will have poor performance, but it will "work."


Labb asked which is "better."  That's a wrong question, but the "right answer" is to explain when each approach is typically used.

   The resistor gets used when the change of screen current from idle to full-power is small.  That keeps screen volts constant-enough to get proper performance from the output tubes.

   The choke gets used when the change of screen current from idle to full-power is large.  That keeps screen volts constant-enough to get proper performance from the output tubes.

    There are ways to avoid using a choke, but there are also tubes that pull less screen current than other alternatives.

    There are also times when people intentionally insert high resistance in the screen circuit to induce sag & reduce total power output.

[jewishjay]

I recently built a low watt single ended Hi Fi stereo. The sound was great, but the hum was unbearable. I tried all the "tricks" to fix it - shielded grid wires, elevated heater CT, ground schemes.....

Nothing stopped the hum until I replaced the first power resistor for a small choke.

Now it sounds great and theres zero hum. You cant even tell its running until the music starts.

[GlideOn]

If you're using a tube rectifier in a higher wattage or Class AB amp, you have other inherent problems...

What 'other inherent problems'?

May mean 'performance characteristics'?
 
Which can be desirable or not.

Nah, let him answer.

You have another point of failure, mostly.
 
And a terrible selection of neither reliable nor affordable rectifier tubes available today.

By all means, have an amp that's fully valve, fully choked. Those don't come cheap these days. Your tech will certainly be very happy!

Is having all of that really the "best" solution?

Is having an electric car "better" than a small block Chevy V8?

$40 choke or $10 cap.

What were we talking about again??? 

[Willabe]

If you're using a tube rectifier in a higher wattage or Class AB amp, you have other inherent problems...

What 'other inherent problems'?

May mean 'performance characteristics'?
 
Which can be desirable or not.

Nah, let him answer.

You have another point of failure, mostly.
 
And a terrible selection of neither reliable nor affordable rectifier tubes available today.

By all means, have an amp that's fully valve, fully choked. Those don't come cheap these days. Your tech will certainly be very happy!

Is having all of that really the "best" solution?

Is having an electric car "better" than a small block Chevy V8?

$40 choke or $10 cap.

What were we talking about again??? 

Not buying it.

If all the new rectifier tubes were bad, we'd see it here.

[AlNewman]

You have another point of failure, mostly.
 
And a terrible selection of neither reliable nor affordable rectifier tubes available today.

By all means, have an amp that's fully valve, fully choked. Those don't come cheap these days. Your tech will certainly be very happy!

Is having all of that really the "best" solution?

Is having an electric car "better" than a small block Chevy V8?

$40 choke or $10 cap.

That's actually a poor analogy.  They both need caps to do their job.
Think of the choke (or resistor) more of a transmission than the engine. 
You have a 4 speed VW bug (resistor) VS an 18 speed tractor trailer (choke).  Pick your poison.

[GlideOn]

You have another point of failure, mostly.
 
And a terrible selection of neither reliable nor affordable rectifier tubes available today.

By all means, have an amp that's fully valve, fully choked. Those don't come cheap these days. Your tech will certainly be very happy!

Is having all of that really the "best" solution?

Is having an electric car "better" than a small block Chevy V8?

$40 choke or $10 cap.

That's actually a poor analogy.  They both need caps to do their job.
Think of the choke (or resistor) more of a transmission than the engine. 
You have a 4 speed VW bug (resistor) VS an 18 speed tractor trailer (choke).  Pick your poison.

A choke resistor being a basic 4 speed transmission vs a multi-gear transmission choke is a pretty good analogy actually. Choke is objectively better in every respect, just a larger component that takes up valuable space.

How about the large cap being the torque converter or clutch that gets beefed up so it's not subject to wear and tear so easily?

Or like a stout first gear that gets the vehicle up and going from a stop quicker?

Larger first stage cap absolutely helps with noise floor, bass response, headroom and overall usability with little to no downsides. Cheapy.

A valve rectifier for better or worse is absolutely another added bit of complexity and point of failure (unless backup diodes are implemented, but rarely so). I'm not hating on them as they undoubtedly have a feel factor to them in some sporting amps, but they do have a worse reputation than solid state rectifiers.

Like having multiple clutches on a transmission when a single stout clutch will do fine.

I'll pick the most proven and economical option all week, every time and most users who would rather pay their tech instead of DIY repair will more than likely do the same.

But the sum of its' parts always makes it what it is.

I don't think anyone here would argue there's a certain charm to driving a dead stock manual 4 speed, rear engined Volkswagen Beetle. My parents had a Super Beetle for a few years. It didn't go anywhere in a hurry, was clunky to operate, but it had an unmistakable presence and cool factor to it even though in actuality it was a bit of a pain in the ass to live with.