resistor on secondary - 32Ω load

[Guitarist]

On a cheapy amp here, I've added a big 32Ω load resistor to an output jack which just stays there. I'm sure I'm not the first one to think about this but the idea is that then I can (and have) add a Line Out. The amp won't see that high a value, I'm hoping.

This is the same value as Ho uses on his Ultimate Attenuator.

Thoughts?

[Chuck H]

On a cheapy amp here, I've added a big 32Ω load resistor to an output jack which just stays there. I'm sure I'm not the first one to think about this but the idea is that then I can (and have) add a Line Out. The amp won't see that high a value, I'm hoping.

This is the same value as Ho uses on his Ultimate Attenuator.

Thoughts?

It's the same value Ho "use to" use on the Ultimate Attenuator. I mentioned that value in another thread and took a shot for it because the current product lit says that they use a "reactive load" without giving specs. Since resistors don't really qualify as reactive (even inductive ones ) I guess there's been changes in the design since I was last familiar with it. Which would be something like nine years ago.

And 32 ohms seems a little heavy for a couple of reasons. What is the actual output impedance of this amp. If it has a 16 ohm secondary option then a parallel 32 ohm resistive load is soaking up more than 1/3 of the amps wattage!!! Why more than 1/3? Lemme splain it...

The resistive load is "fixed" but speakers present a reactive load with a resonance due to inductance that is really very high in the HF and at the resonant peak. Really, the only areas where a guitar speaker presents the rated load would be in the LF below useful frequencies for guitar amps and in a broad area of the midrange frequencies. Since the resonant peak might be something like a hundred ohms for even an 8 ohm speaker you can see where a parallel 32 ohm resistive load might alter the amps response and tone due to a flattening of the load resistance. This happens in the HF too, but to a slightly lesser degree.

One reason I read for the Ho using a 32 ohm load was that it sort of averaged the actual impedance @ frequency as presented by real world speakers. Other attenuators have done the same. That is, use a higher resistive load than the rated speaker load for a resistive attenuation. The TrainWreck Airbrake does this too (IIRC). So...

Maybe a good idea for an attenuator, but I don't think it's good as a safety load resistor. Certainly the smaller you go, the more protection you have (for tube amps), but I don't think this should come at the expense of amplifier performance in any noteworthy sense.

I think Traynor has used 270 ohm and even 470 ohm resistors as safety resistors permanently fixed in parallel with the speaker jack. I can't claim to know the technical specifics, but I remember reading (here, many years ago) that this is a huge level of protection compared to an actual open load. There was even some math involved demonstrating the extreme voltage spikes of an open load and how even a parallel resistor in the hundreds squelched it into the safe zone.

[g1]

I think Traynor has used 270 ohm and even 470 ohm resistors as safety resistors permanently fixed in parallel with the speaker jack. I can't claim to know the technical specifics, but I remember reading (here, many years ago) that this is a huge level of protection compared to an actual open load. There was even some math involved demonstrating the extreme voltage spikes of an open load and how even a parallel resistor in the hundreds squelched it into the safe zone.

I can't recall who (KG maybe?), somebody on here tried their darndest to damage an OT using a safety as high as 1K I believe. If memory serves, even that did the job. (I recall they were unable to do damage, just uncertain of the exact value)

[Helmholtz]

Really, the only areas where a guitar speaker presents the rated load would be in the LF below useful frequencies for guitar amps and in a broad area of the midrange frequencies.

Below the bass resonance speaker impedance is mostly below rated, approaching the DCR at very low frequencies. Rated speaker impedance is typically specified at 400Hz. In the region from 200Hz to 500Hz it stays close to rated. At 1kHz an 8 Ohm speaker shows about 10 Ohm. In normal operation most of the average program power is produced within a frequency range where speaker impedance doesn't exceed twice its rated value.

While a load of twice the nominal impedance is probably safe, further increasing may reduce tube life. Reason is that high output load reflects to high tube load impedance, strongly increasing average screen dissipation.

[Guitarist]

Thanks for the insights. I'll try the amp out soon I just had to get a proper drill bit to add the Line Out jack.
This particular amp has 4 and 8Ω outs. But Chuck, losing a third power seems ok wit me. The amp has 6v6 outputs so playing with a band, it should still be heard.
I'm not too set on using 32Ω and if a higher value, say 64Ω, could be used then a slightly smaller resistor Wattage size as well.
Traynor: 180 ohm 20 W
BuilderX 270 ohm 5W
Victoria 470 ohm 2 W

I suppose there may be a sweet spot which doesn't stray too far from 16Ω as we want OT and some attempt at frequency curve preservation. I estimate that a 64Ω load would need a 20 (15W dissipation) Watt resistor and a 32Ω a 25 watt for a 50 Watt amp. But math isn't my strong suit
I'll also add a string of three 1N4007 from the OT to ground.
Now to figure out the Line out values to go Line In on an the interface.

[Helmholtz]

As long as you are using speakers, there is no need for an additional load resistor for line-out. If you don't use speakers, please make sure that the dummy load resistance is not higher than twice the nominal output impedance, otherwise your power tube screens are at risk.

Traynor: 180 ohm 20 W
BuilderX 270 ohm 5W
Victoria 470 ohm 2 W

I don't understand this list.

[Chuck H]

The risk to the OT from an open load is arching between windings. You basically need a resistance that is sure to be lower than the magnet wire insulation resistance. Probably not a tall order where you need to get really low in value. I think you might just as well use something like 150R (if a lower number pleases you) or 220R and be sure there's no noteworthy change to how the amp "sees" the speaker load.

[Chuck H]

As long as you are using speakers, there is no need for an additional load resistor for line-out. If you don't use speakers, please make sure that the dummy load resistance is not higher than twice the nominal output impedance, otherwise your power tube screens are at risk.



I don't understand this list.

The circuit in question is a safety resistor parallel to the speaker load. In the event that no speaker is plugged in it should offer some protection to the OT.

The list indicates values used by other manufacturers for this circuit.

The line out was only just mentioned and is a different topic.

[Helmholtz]

The risk to the OT from an open load is arching between windings. You basically need a resistance that is sure to be lower than the magnet wire insulation resistance. Probably not a tall order where you need to get really low in value. I think you might just as well use something like 150R (if a lower number pleases you) or 220R and be sure there's no noteworthy change to how the amp "sees" the speaker load.

I am not speaking of OT risk, but of melting power tube screens at full power.

[Chuck H]

I think we got out of step with our posts. The safety resistor isn't the only load. It is hardwired across the speaker jack and a normal speaker is still supposed to be plugged in. The discussion of attenuators that use a 32 ohm load was a side topic.

[Helmholtz]

The circuit in question is a safety resistor parallel to the speaker load. In the event that no speaker is plugged in it should offer some protection to the OT.

The list indicates values used by other manufacturers for this circuit.

The line out was only just mentioned and is a different topic.

Ok, maybe I misunderstood. These resistors may protect the OT from excessive primary voltages but will not protect the screens at large signals.

[Guitarist]

This amp will likely always have that resistor internally attached. It's been stated by Chuck that it will only lose 1/3 power which is just fine. I'm also now adding higher screen resistors of 2K so that'll also help. Diodes too on the plates.

I've been practicing/recording with various amp-simulators and for my little studio this modest little amp promises to really help with this process.

[Helmholtz]

This amp will likely always have that resistor internally attached. It's been stated by Chuck that it will only lose 1/3 power which is just fine. I'm also now adding higher screen resistors of 2K so that'll also help. Diodes too on the plates.

I've been practicing/recording with various amp-simulators and for my little studio this modest little amp promises to really help with this process.

But why then do you need this additional load resistor? Do you need to unplug speakers during operation?

BTW, without input signal standard shorting speaker jacks are safe enough.

[Guitarist]

Yes exactly, no speakers attached. I'll use speaker emulation impulses. Who knows, this may suck but I'm trying to get some better tones. ..
Good to know about the shorting jacks, thanks.

[Chuck H]

JM2C on the matter...

If this amp is going to be used with several different speaker loads and simulated loads in a small studio environment the odds that it could be accidentally operated without a load plugged in are higher than average. And in fact it's my opinion that a lot of the bad rep attenuators and Marshall heads get for being bad for each other (not so much now, but it use to be a thing) is for the same reasons. When there are more things involved with plugging in the load more mistakes get made. So in this instance a safety resistor is a good idea.

And, no, it wouldn't help with screen dissipation. But just because a seat belt won't protect you from fire resulting from a collision doesn't mean you shouldn't wear one