Hi, I've designed a simple 100W power attenuator for a guitar amp. When doing some calculations, I realised that at 20,000Hz the total impedance is reduced to 6.8ohm from 8ohm due to the capacitor high pass filter. Will this be a significant issue or is there a way of adding a high pass filter without affecting the overall impedance i.e. a resistor with the capacitor? The amp and speaker impedance are both 8ohm.

 

One question.
Why did you decide to attenuate the output signal from the power amp output, instead of just simply attenuated the input signal before it reaches the power amplifier?

 

One question.
Why did you decide to attenuate the output signal from the power amp output, instead of just simply attenuated the input signal before it reaches the power amplifier?

This is attenuating from the speaker output to the speaker. It's for use on a valve amp so to achieve a good tone the amp is turned up.

 

The TS is trying to generate more distortion from the amp.

What's the purpose of the high pass filter here?

 

Fletcher Munson curves mean that the perceived loudness of higher frequency will be less when attenuated. The filter is to allow high frequencies to pass without attenuation.

 

Hi, I've designed a simple 100W power attenuator for a guitar amp. When doing some calculations, I realised that at 20,000Hz the total impedance is reduced to 6.8ohm from 8ohm due to the capacitor high pass filter. Will this be a significant issue or is there a way of adding a high pass filter without affecting the overall impedance i.e. a resistor with the capacitor? The amp and speaker impedance are both 8ohm.

Hi,

Not sure exactly what you are doing but 6.8 ohms is not far from 7 ohms, and 7.5 ohms is not far from 8 ohms, so add a 0.5 ohm resistor in series with your attenuation circuit and you'll be close to 8 ohms. For the other frequencies you imply that it is 8 ohms, so for them it will then be 8.5 ohms which isnt bad either.

 

your calculations are correct but speaker impedance is measured at 400Hz. This is an L Pad attenuator.

Have you tried it yet? Hope you have some good air flow going to it.

 

Fletcher Munson curves mean that the perceived loudness of higher frequency will be less when attenuated.

No, they don't.

First, the difference between midrange and high frequency sensitivity is large enough to matter only at very low loudness levels; for the "normal" range of volumes, the difference is effectively constant.

Second, at high volume levels the difference actually reverses.

Third, the curves were developed in the 1930's, when the world was a quieter place both physically and acoustically. Later experiments indicate that where you live (urban vs. rural) has a huge effect on your perception.

Fourth, your hearing changes based on what you are hearing. This was a problem with the original data - there were no controls on things as simple as whether the windows were up or down in the car on the way to the test, and we now know that something like that has a direct effect on the test results. Whatever someone thinks of the midrange/treble balance in the audio at the start, it will be different 10 minutes later.

ak

ps. did 6 years designing instrumentation for a psycho-acoustics lab.

 

No, they don't.

First, the difference between midrange and high frequency sensitivity is large enough to matter only at very low loudness levels; for the "normal" range of volumes, the difference is effectively constant.

Second, at high volume levels the difference actually reverses.

Third, the curves were developed in the 1930's, when the world was a quieter place both physically and acoustically. Later experiments indicate that where you live (urban vs. rural) has a huge effect on your perception.

Fourth, your hearing changes based on what you are hearing. This was a problem with the original data - there were no controls on things as simple as whether the windows were up or down in the car on the way to the test, and we now know that something like that has a direct effect on the test results. Whatever someone thinks of the midrange/treble balance in the audio at the start, it will be different 10 minutes later.

ak

ps. did 6 years designing instrumentation for a psycho-acoustics lab.

Yeah actually you're right, in fact looked at it again and if anything the high frequency would be more pronounced. Weird as tutorials for attenuators have said that the high frequencies were lost when they turned up the attenuation, perhaps this is more an electronics rather than psychoacoustics thing, interesting. Maybe I'll leave it out and see how it sounds for myself, I'll do some frequency analysis to see what effect it has.

Cheers AnalogKid

 

your calculations are correct but speaker impedance is measured at 400Hz. This is an L Pad attenuator.

Have you tried it yet? Hope you have some good air flow going to it.

Yeah I was just seeing what effect a 20k audio signal would have on the reactance of capacitor and therefore overall impedance. Yeah I designed it myself, pretty much just a voltage divider (though yeah know as an Lpad, because of the shape I guess).

Just a concept but will have thermal paste on resistors and lots of slots in the aluminium case!

Cheers Wolframore

 

Hi,

Not sure exactly what you are doing but 6.8 ohms is not far from 7 ohms, and 7.5 ohms is not far from 8 ohms, so add a 0.5 ohm resistor in series with your attenuation circuit and you'll be close to 8 ohms. For the other frequencies you imply that it is 8 ohms, so for them it will then be 8.5 ohms which isnt bad either.

That's not a bad idea at all actually. It basically tends toward 8ohm as frequency decreases.

 

The TS is trying to generate more distortion from the amp.

What's the purpose of the high pass filter here?

For some reason lots of people have said that attenuating reduces the high frequencies so trying to let these bypass the attenuation through the cap.

 

To me attenuating always feels thinner as in the bass is missing. Either way ak is right. I hate this subjective discussion. It’s like recording or mixing with the NS10s to make sure we add enough bass to the mix and scoop the mids. Did you also notice bass takes way more power to balance the mix? And you’re right there are a lot of psychological effects. I’ve seen people use devices with bunch of lights to fix mixes for artists saying something is missing when all is well. Those devices did nothing.

I have a 5 watt tube amp that sounds amazing to record.

 

To me attenuating always feels thinner as in the bass is missing. Either way ak is right. I hate this subjective discussion. It’s like recording or mixing with the NS10s to make sure we add enough bass to the mix and scoop the mids. Did you also notice bass takes way more power to balance the mix? And you’re right there are a lot of psychological effects. I’ve seen people use devices with bunch of lights to fix mixes for artists saying something is missing when all is well. Those devices did nothing.

I have a 5 watt tube amp that sounds amazing to record.

Exactly
I'll leave it for now and see what it sounds like.

I always have a DFA(does fuck all) knob assigned in live sound which the band notice the difference when they want something tweaked.

 

To me attenuating always feels thinner as in the bass is missing. Either way ak is right. I hate this subjective discussion. It’s like recording or mixing with the NS10s to make sure we add enough bass to the mix and scoop the mids. Did you also notice bass takes way more power to balance the mix? And you’re right there are a lot of psychological effects. I’ve seen people use devices with bunch of lights to fix mixes for artists saying something is missing when all is well. Those devices did nothing.

I have a 5 watt tube amp that sounds amazing to record.

5 watts is the better option. It's for a friend who wants to practice with a 100W tube amp, though it would be a cool project.

 

Even my 60 watt gets too loud to overdrive it properly. It is good to practice with what you plan to use. Forget about running my 120 watt head. The 5 watt is made by Blackstar who are expats from Marshall. I got it donated to me cause there was an issue with a mosfet. It was a design flaw on their first runs. I moved the MOSFETs and put heat sinks on them. It runs ECC83 preamp tube, a 12BH7 power tube at 370v. It’s a nice sounding amp.