Power attenuation by parallel resistor

[bpdude]

I would like to make a power attenuator for a friend of mine who uses a tube amp.
My idea is to put a power resistor parallel to the cabinet, and to use an output on the amp that is meant for the resulting impedance. The resistor would be there to act as an additional load so the amp wouldn't suffer, and the "attenuation" would actually be done by the output transformer.
(So if I had a 16ohm cabinet, I could connect it to the 8ohm output together with a 16ohm resistor)
I find this to be a potentially better idea than an L/T pad because it will not alter the damping factor of the amp by much, so it won't change the tone, only the volume.


The reason why I started this thread is because I can't find a place on the net where this has been done before, so I think all this might be too good, too easy to be true.

Any opinions?

[JohnH]

I think it would work, but it would a only be a small volume reduction of 3db.

You'd need resistors, in a series and/or parallel chain totalling 16 Ohms, that have a total power rating comfortanly above their half share of the power. eg io soak up 50% of a 50W amp, I reckon youd want 50W of resistors. They need to be mounted with good air circulation all round, such as on a tag strip in a vented metal box. The metal box should not be connected to either of the output wires.

John

[gdgross]

I think it is too good to be true.

I resistor doesn't look like a speaker as far as impedance. Although your speaker cab might be rated for 8 or 16 ohms, in reality the actual impedance varies quite a bit with frequency. As a result, your speakers and your resistor (network) will not share power equally across all frequencies, and this can't help but affect your tone.

Regardless, try it anyway. Maybe it will affect your tone in a good way!

[sumgai]

bippy,

gross is somewhat correct, speakers vary in their impedance across the frequency spectrum. However, in all that variance, the audible power output does not vary nearly as much. This action is due to the amount of current flow through the speaker as a function of the impedance. Any decent power supply will easily accomodate the normally-seen wide variances in impedance.

An old trick used to protect speaker is to place a 100watt (120v) bulb in series with the speaker in question. This may not reduce the volume very much by itself, but if you were to string several of them together...... Series-connectioned bulbs will introduce additional resistance, and you may find the ideal effect, who knows.

If you use colored bulbs instead of white, you'd have an old-fashioned color organ!

If you do experiment as suggested by John, you can alter the resistance that you place in parallel with the speaker. That will let you feed only a given amount of power to the speaker, the rest being soaked up by the lower resistance of your "foot warmer". Ventilation, meaning getting rid of the heat build-up, will be the order of the day!

The forumula for calculating all of this is:

R_s X R_added / R_s + R_added

where "s" is the speaker's impedance, and "added" is the extra resistance you add in parallel to the load.

HTH



sumgai

[bpdude]

Thanks!

I want to avoid any series resistance, not to change the impedance seen by the cabinet. As I understand it, the cab will care about the voltage across its pins, the damping factor, and that's about it.


Other than that, what I want to achieve is not saturation at bedroom levels, but an increase of output tube overload at the same level for gigs and rehearsals. 3dB (I guess 6 dB by going from 16 ohms to 4) would be enough to, you know, go above the top; push the envelope; turn it up past 10

So thank you all for the good comments. I just needed some reassurance that I'm not doing something completely stupid.

[JohnH]

Yes, if you have a 4Ohm output on the amp, you can use a 5.3Ohm resistor network in parallel with the 16Ohm cab to get 4 Ohms. It would be neat to have both 16 and 5.3 ohm options available, to select -3db or --6db.

It may still affect tone, since its a resistive load rather than an inductve one like an all speaker group, but I think its a good thing to try - and lets hear how it sounds.

John