I really like this approach with the two adjustable resistors. I think I might build one like this too.
It won't react like an actual speaker load which will change impedance depending on frequency, but at least it presents a more consistent load to the amp than a standard airbrake.
Has anyone taken into consideration that the load also has some loading from the speaker cab? I'm not the math wiz, but I wonder how that all fits into the equation.
Yes we have. The total impedances I mentioned are what the amp "sees" with the attenuator and cab connected. By using a combination of a series and a shunt resistor it is possible to keep that load quite constant.
Got back from my holiday yesterday, so I thought I would put the 4 Ohm Airbrake together today. No problems with that, takes just a little more effort than a regular Airbrake to build. Just check everything with your digital multimeter to be certain you connected everything the right way.
It sounds fantastic. Every bit as transparent as the regular Airbrake, with the same pros and cons: Nice and pretty transparent attenuation for the first couple of clicks. After three you start to notice the loss of high end (and I guess also the sound pressure, psycho-acoustics thing).
The bedroom attenuation at max sounds pretty fizzy, as does the regular Airbrake. This is in part due to the 0.100uF cap used to preserve some of the highs. Might be useful to try out different values, or to lose the cap al together.
The thing this version has over the regular one is the constant impedance. I calculated the attenuation steps: 0dB attenuation, -3.1dB, -5.2dB, -7.2dB, -9.2dB, -10.94dB and -23 dB at max attenuation. The total impedance the amps sees at these steps (measured with cab connected) is 4.1 Ohms, 3.6 Ohms, 3.8 Ohms, 3.7 Ohms, 3.9 Ohms, 4.2 Ohms and 5.3 Ohms at max attenuation (rheostat maxed).
Downside to this design is that I have the impression the shunt resistor, which is also variable in this design, gets much hotter at max attenuation than the original design does. That is not strange, as at max attenuation only about half of the 10 Ohm, 100 Watt resistor is used. I'll have to do some more testing to see how it holds up after prolonged playing. As I am mostly going to use this at gigs at the first 2 or 3 attenuation steps, I don't think it will be a problem.
To conclude: it is very easy to build a good, reasonably transparant attenuator, while keeping the load your amp sees almost completely constant and at the right value. Recommended modification to the Airbrake design, as far as I'm concerned.
Well I just built mine, although it's suited for 8 ohms.
I just realized after building mine that the wattage rating for the adjustable resistors goes down proportionally when you put the tap at a lower value. Mine has been okay so far since I used 100W resistors and I have 6v6's in my express, but if I used a 50W amp, I would be overworking the resistors at some attenuation settings. Actually, if i run it fully cranked I might still have to be careful.
It sounds great and presents a nearly constant impedance, but I thought I'd just give the warning about the possibility of overpowering the resistors.
Yes, that's the exact same thing I noticed: Resistors get considerably hotter than with the regular Airbrake design.
I haven't played through the modified design longer than 30 min at a time. It didn't give any problems at that.
I agree: be extra careful not to overpower the resistors.
This is probably one of the reasons KF designed it the way he did.
You can do it, but you'd have to change the values appropriately.
I attached a resistor calculator i made in excel. the first page shows the stock airbrake values and lets you plug in your amp wattage and speaker impedance to see what values a stock airbrake would present to your amp in steps 2-5 of attenuation. Step 1 is no attenuation and is just the speaker plugged into the amp so I did not inclued it.
The second page calculates ideal resistor values based on the amp wattage and expected impedance. The values it shows are based on reducing speaker output power by half with each step and then adding a rheostat that will take it down to -30dB.
Just remember that tube amps are generally rated in clean watts. In order to make it safe, you should bump up the resistor power ratings by a safety factor of 2 or 3, I'm not exactly sure what's safe for resistors. Maybe someone else can chime in on what minimum safety factor would guarantee success.
You do not have the required permissions to view the files attached to this post.
I think I might build one like this too.