Choke/resistor switch?

[daz]

I like using a resistor instead of a choke sometimes in my marshall 50 style build, and i wanted to see if anyone could tell me whether there are any reasons why that wouldn't be a good idea to have it switchable and whether there would likely be any nasty pops that would require turning the amp off to switch it. The choke or resistor is after the OT primary and before the screens. In other words between the 2 sides of the 32/32uf can. Thanks.

[Old Tele man]

...you'll get a noticeable change in screen (Vs) voltage between the two settings, which means "rebiasing" (unless amp is cathode-biased) between the two settings:

A) CHOKE, having much lower DC-resistance (typically, <100 ohms), has a lower voltage "drop" for a given current and so will provide a HIGHER Vs voltage.

B) RESISTOR, having greater resistance (typically, ~1K ohm), has a greater voltage "drop" for a given current and so will provide a LOWER Vs voltage...and, subsequent circuits (ie: "down stream" tubes) will ALSO be getting lower voltages too.

...a 'higher' Vs will need additional negative bias voltage to maintain the same idle current; and, conversely, a 'lower' Vs will need less negative bias voltage to maintain the same idle current.

...if you're doing this just to "see/hear" what happens, it'll be OK so long as you re-bias appropriately. Wouldn't recommend it be intended for normal everyday "flipping" while playing usage, however.

[daz]

well, the thing is that the bias isn'y far off and i already figured i could use a DPDT switch and have it add in a set resistor of whatever value i determine would keep the bias the same. But what i wanted to know is if there was any reason why it wouldn't be a good idea or whether it would be ok to switch it while the amp is on or at least with standby one. i don't intend to use it as i'm playing. i just intend it to be a way to change the feel and sound when i'm in the mood for something a bit different or to tune a room.

[daz]

I have another question here. As i said i can use a DPDT to also change the bias when i switch from choke to resistor. But it's pretty close as is and if it were a bit closer i wouldn't even have to bother. So the question is this....i'm using a 1k resistor right now. How do i determine how low i can go on the value, and besides keeping the bias in range what might happen tone-wise when using a much lower value than 1k?

[R.G.]

A resistor has equal impedance at all frequencies (ideally).
A choke has low, almost zero impedance at DC and it rises steadily.

The important differences in your use is that
(a) at DC, the resistor drops more voltage than the choke. How much more? Measure the choke's DC resistance. It's probably under 50 ohms. The 1K resistor is just 1K. So the resistor drops 20 times the DC voltage of the choke. That's what makes the DC change, changes the screen voltage and hence the "gain" of the tube.
(b) at 120Hz, the ripple frequency, your resistor is still 1K. The choke is Xl=2*pi*F*L = 753 ohms per henry plus the DC resistance. If your choke is more than 1.37H, then the choke has a higher impedance at 120Hz, and is better, when combined with the decoupling cap, than the resistor at reducing AC ripple.
(c) at signal frequencies above 120Hz, the choke gets ever-better at preventing coupling through the power supply of signal frequencies that could cause oscillation.

[daz]

You're over my head a bit there, but if it helps, the choke is a 20H which i chose because when i built it i was going to copy a matchless which uses a 20H. In any case, can you determine from what i've said what value of R i should use? if not, what else do you need to know? As for hum, theres no difference right now with the 1k over the choke. It's a very quiet amp to begin with tho.

[Old Tele man]

...to continue RG's excellent explanation, and to answer your question of "...what size resistor?" you simply use the XL-equation that RG provided and solve for the equivalent impedance/resistance (R = Z = XL) value:

Z = 2*PI*Hz*L = 15,079.6 ~ 15.1K-ohms (including the 50-ohms winding resistance)

where:
PI = 3.14159...
Hz = 120 cycles-per-second
L = 20 Henries

...thus, a 15.1K-ohm series filter resistor will have the same filtering ability at 120Hz (the fullwave rectification frequency) as the 20H choke, assuming no change in the before/after filter capacitors.

...however, a 15K resistance is gonna "drop" a lot more voltage than will the 20H/50-ohm choke, which will VERY noticeably reduce screen voltage (Vs) -- and, likewise, tube "gain" -- and so will require re-adjustment of bias voltage to maintain the same idle current. Because plate voltage (Vp) is taken BEFORE the choke, it should not change much (but might, slightly).

...with rectification, the ripple voltage contains a FUNDAMENTAL (or DC-component) and HARMONIC frequencies (ie: 120Hz, 240Hz, 360Hz, etc.). A resistor affects all of them the same, but a choke, with it's essentially zero-resistance to DC will have almost NO affect on the DC-component (except it's 50-ohm winding resistance). Since the DC-component from a fullwave rectifier is 63.7% of the peak voltage, the choke will "drop" only a VERY small portion of this DC-value; while the constant-valued resistor will "drop" a much LARGER portion of this DC-value.

...a "perfect" filter should have NO opposition to the FUNDAMENTAL, but have infinite opposition to ALL frequencies above the filter's cutoff frequency, which for fullwave rectification is 120Hz.

[daz]

I may be missing something. After all, this is for the most part over my head. But of all the amps i've looked at that use a R instead of a choke, the largest R i have seen was 1.5k and the smallest around 330 ohms. Most are under 1k. So 15k seems awful high. Are you sure 15k is right?

[Enzo]

It depends on what you are asking. Do you want the resistor to have the same regulating effect as the choke? Or do you want the voltages to be the same when selecting resistor or choke? Read the whole explanation, don't just look for a resistance number. he explained that 15k would drop a ton of voltage compared to the choke, but it would filter the same amount If you use a smaller resistor to keep the voltage up, then the filter action will be reduced. There is no free lunch.

It will be a compromise, but if the plan is to make the circuits more or less equivalent in their function, what then is the point of a switch?

The voltage dropped across a resistor is a direct function of the current through it. The resistor values in an amp are designed to set the desired voltage for each stage powered. If you want 480 volts on the power tube plates, and 460 on the screens, then you select a resistor to drop 20 volts at the current running through it. In that example, in a typical amp, there would be the screen current PLUS all the preamp tube currents. Then from the screen node (node just means a step in the B+ string) to the next node - the phase inverter for example - maybe we want to drop another 100 volts. So we select a resistor to drop 100 volts at the current of the PI plus the tubes earlier in the circuit. The power supply is designed to provide the voltages we want or need. Saying Fender uses this or Marshall uses that is only part of the story.

[R.G.]

I may be missing something. After all, this is for the most part over my head. But of all the amps i've looked at that use a R instead of a choke, the largest R i have seen was 1.5k and the smallest around 330 ohms. Most are under 1k. So 15k seems awful high. Are you sure 15k is right?

For the conditions he sets up yes. And as Enzo points out, you have two different things going on. One is dropping voltage, the other is filtering.

For a resistor, DC voltage drop and filtering are irretrievably tied together. DC drop is current times resistance. Filtering rolloff is 1/R.

For an inductor, you can get ever-increasing filtering effectiveness with almost no DC loss; but you have to pay for this by buying an expensive inductor. The reason amps went with inductors in the first place instead of a cheaper resistor is that they could get the screen voltage up quite high (for high gain in the output tubes) because of the low DC losses, and still have the screen voltage filtered of power supply hum at the same time.

Amps which use resistors there put in 1K or 1.5K because that's all the DC voltage drop they can stand. They CAN'T us 15K, or the DC voltage drop would be too much and the screen voltage would be too low. That's why they stay with small resistors. They have to.

But yes, a 20H inductor filters like a 15K resistor at 120Hz, the frequency of power supply hum. That is correct.

[daz]

I see now. I wasn't exactly sure why chokes are used in some amps and not in others. so basically the effect i'm feeling is the voltage drop. So i assume if i left the choke in and added the R at the beginning of the B+ like some do to create sag, it would be basically the same as no choke and the R in it's place except the filtering wouldn't be as good. Although a sag R is different in that it's before the OT and therefore drops the voltages there too. And maybe thats why when i tried a sag R before i didn't care for it but with the R as filter with no choke i now like it. Because it's AFTER the OT, ya think?

[Gregg]

I'm not very much into theory but if you check the schematics of some "inductorless" amps like Marshall, Peavey etc. you'll notice that the resistor in question is usually a 100-470 Ohm/5-10W resistor.

[daz]

But the thing is, i'm not sure i want to go that low if the tone i'm getting becomes closer to sounding like a choke as has been suggested. By the way, every marshall i've owned had a choke. What models are you referring to. I'd be interested to look at those schematics.

[Gregg]

What models are you referring to. I'd be interested to look at those schematics.

Check out JCM900s, TSL60/100, the last 4 channel JVM410 etc. Most of the Peavey amps have a resistor as well.

[daz]

Thanks. yeah, i went to schematic heaven and it hit me when i saw the 900 that i had one and as far as i recall it had no choke. So i looked at the schematic and sure enough a 100 ohm. So i tried a 270 which was the lowest value i had, and i will try a 100 ohm today. I don't know if it's placebo effect or what, but the amp seems to have more clear and sweet sparkle in the cleans the lower i go on the R value. Anyways, the amp is very quiet and now i'm having a hard time understanding the need for a choke at all unless the particular circuit is prone to being noisy. But for mine i now see absolutely no need for a choke, and i also don't think i'm going to make it switchable because i think it sounds and feels much better like this. Of course i know how things like this go, so i will add a temporary switch just to be sure the choke doesn't offer an alternative sound/feel that i'd also like to have available. If not, my amp will be losing about 3 Lbs ! (yes, it actually DOES weigh that much according to hammond's specs. Actually 3.25Lbs to be exact)