Reactive Load Question

[Reeltarded]

This is just going to make Martin more angry since we aren't mathing in a circuit.

We are just seeing what it does at the end. It appears to have a similar curve to a cabinet... I think it's ok to know what it looks like, but it's not important to how it feels.. here's why..

I never play without a load on every amp, but I load amps to channelize them and they end up being rammed through 4x12 cabinets. I would not want frequency response before the cabinets to mimic cabinet frequency response, but I would like the amp to feel like anything but a brick wall..

So. Your load sounds like a cabinet to our eyes. If it feels like one under your fingers: good enough.

No amount of anything we do to these things is ever going to be a substitute for an amp pushing a speaker in a baffle until the room moves enough to become part of the system. It's a constant like any other physical truth in "the" reality.

[Cathode Ray]

This is just going to make Martin more angry since we aren't mathing in a circuit.

We are just seeing what it does at the end. It appears to have a similar curve to a cabinet... I think it's ok to know what it looks like, but it's not important to how it feels.. here's why..

I never play without a load on every amp, but I load amps to channelize them and they end up being rammed through 4x12 cabinets. I would not want frequency response before the cabinets to mimic cabinet frequency response, but I would like the amp to feel like anything but a brick wall..

So. Your load sounds like a cabinet to our eyes. If it feels like one under your fingers: good enough.

No amount of anything we do to these things is ever going to be a substitute for an amp pushing a speaker in a baffle until the room moves enough to become part of the system. It's a constant like any other physical truth in "the" reality.

I upped the response time - exact same track here being analyzed :



My favorite way to record is with this reactive load connected to one 16 Ohm tap off the amp, and the 4x12 cab connected to the other... then mic the cab with an SM57 and mix the two signal in the DAW.

[martin manning]

Didn’t see your PM due to full inbox...

Here are some detailed instructions for measuring speaker load impedance using only a multimeter and a signal source. The goal is to determine the load impedance as a function of frequency, and to determine the frequency of the resonant peak and its magnitude. Calculations and plotting of results is easily done using spreadsheet software.
1. A low-level signal (100 mV) which can be varied in frequency is required, along with an amplifier that produces a clean sine wave at its output. A mobile phone app works well for the signal source, and can be connected directly, or wirelessly through a Bluetooth adapter. A 3.5mm to 1/4” phone plug adapter may also be required.
2. Connect the load through a series resistor (Rs), the value of which is somewhere around the expected load impedance. The voltage measured across this resistor VRs [V] will be used to calculate the current flowing through the load IRs = VRs/Rs [A]. An attenuator or speaker load simulator may have a series resistor at its input which may be used for this purpose.
3. Turn the signal generator and the amplifier on, and adjust the amplifier’s volume so that the voltage measured across the load is one or two volts, at a frequency which is well above the resonant peak, say 1kHz.
4. Measure the voltage across Rs and the voltage across the load Vload [V], input to ground, and record these two values.
5. Repeat the process 4., reducing the frequency by half such time and recording the two voltages, until the frequency is well below the expected resonance peak, say 30 Hz.
6. Finally, increase the frequency and find the frequency where the voltage across Rs is minimized. This will be the frequency of the resonant peak.
7. For each set of measurements calculate the load impedance as Zload [ohms]= Vload/Iload = Vload x Rs/VRs
8. Plot Zload vs. frequency (log scale) to see the impedance characteristic.
MPManning 2018.10.21

[Cathode Ray]

Didn’t see your PM due to full inbox...

Here are some detailed instructions for measuring speaker load impedance using only a multimeter and a signal source. The goal is to determine the load impedance as a function of frequency, and to determine the frequency of the resonant peak and its magnitude. Calculations and plotting of results is easily done using spreadsheet software.
1. A low-level signal (100 mV) which can be varied in frequency is required, along with an amplifier that produces a clean sine wave at its output. A mobile phone app works well for the signal source, and can be connected directly, or wirelessly through a Bluetooth adapter. A 3.5mm to 1/4” phone plug adapter may also be required.
2. Connect the load through a series resistor (Rs), the value of which is somewhere around the expected load impedance. The voltage measured across this resistor VRs [V] will be used to calculate the current flowing through the load IRs = VRs/Rs [A]. An attenuator or speaker load simulator may have a series resistor at its input which may be used for this purpose.
3. Turn the signal generator and the amplifier on, and adjust the amplifier’s volume so that the voltage measured across the load is one or two volts, at a frequency which is well above the resonant peak, say 1kHz.
4. Measure the voltage across Rs and the voltage across the load Vload [V], input to ground, and record these two values.
5. Repeat the process 4., reducing the frequency by half such time and recording the two voltages, until the frequency is well below the expected resonance peak, say 30 Hz.
6. Finally, increase the frequency and find the frequency where the voltage across Rs is minimized. This will be the frequency of the resonant peak.
7. For each set of measurements calculate the load impedance as Zload [ohms]= Vload/Iload = Vload x Rs/VRs
8. Plot Zload vs. frequency (log scale) to see the impedance characteristic.
MPManning 2018.10.21

Thank you, Martin.

[martin manning]

Get the measurements and let’s see what you come up with!

[Cathode Ray]

Get the measurements and let’s see what you come up with!

Curious to see how close the results are to your software model result.

[Cathode Ray]

Hey Martin,

So I have this experiment setup exactly the way you instructed.

I have my iPhone play a 1kHz tone into the input of my amp thru an adapter cable.

I have the amp set clean, and the 16 Ohm tap going into the reactive load.

I turned the amp master vol up until I see 2 VAC on the meter (measuring at the input jack inside the load box).

Something really weird (and unexpected) is that when I turn the amp up enough to get 2VAC, I can actually hear the 1kHz tone coming from the components inside the load box.

Anyway, when trying to get a reading from the hot (tip) on the input jack to the other side of those 2 power resistors, I see almost no deflection at all on my multi-meter.

Possibly my meter is not sensitive enough ?

The lowest VAC setting on it is 10 VAC.


Thanks!

[martin manning]

At 1kHz you should see most of the 2V rms input dropped across the 16.5 ohm series resistance, maybe 1.95 V. At the resonance peak, most of the voltage will be dropped across the reactive components, and the voltage drop across the series resistance will be reduced to around 0.35V (350 mV), so you will need a meter that can measure in that range.

It’s not surprising that you can hear the components singing. Output transformers will do that too.

[Cathode Ray]

At 1kHz you should see most of the 2V rms input dropped across the 16.5 ohm series resistance, maybe 1.95 V. At the resonance peak, most of the voltage will be dropped across the reactive components, and the voltage drop across the series resistance will be reduced to around 0.35V (350 mV), so you will need a meter that can measure in that range.

It’s not surprising that you can hear the components singing. Output transformers will do that too.

Off to Amazon to order a better quality meter...

Any recommendations ?

[martin manning]

Fluke meters are considered by most to be the gold standard. Of those, a model 115 is a good meter for general purpose use.

[pompeiisneaks]

I've got the fluke 117 and love it, it's only a bit more than the 115.

~Phil

[martin manning]

Pretty complete comparison of Fluke meters here: http://www.transcat.com/media/pdf/Fluke ... -Chart.pdf

[Cathode Ray]

Pretty complete comparison of Fluke meters here: http://www.transcat.com/media/pdf/Fluke ... -Chart.pdf

Ordered the Fluke 115. Will be here Friday.

Jumping ahead a little bit, let's say the resonant peak isn't where it should be, before I start replacing components wouldn't it be possible to simply remove small amounts of wire from the inductor(s) to alter the resonant peak ?


Thanks

[martin manning]

Jumping ahead a little bit, let's say the resonant peak isn't where it should be, before I start replacing components wouldn't it be possible to simply remove small amounts of wire from the inductor(s) to alter the resonant peak ?

I wouldn’t go there. It’d be easier to adjust the value of the 200u capacitor to move the peak. To make up an alternate value, paralleling two caps would be the way to go. For example if the peak is at 80Hz and you want to move it to 90Hz, the new capacitance would be smaller by the square of the frequency ratio, a factor of (80/90)^2, so 158u. You’d have to measure the caps too, since the tolerances are quite high.

[Cathode Ray]

Jumping ahead a little bit, let's say the resonant peak isn't where it should be, before I start replacing components wouldn't it be possible to simply remove small amounts of wire from the inductor(s) to alter the resonant peak ?

I wouldn’t go there. It’d be easier to adjust the value of the 200u capacitor to move the peak. To make up an alternate value, paralleling two caps would be the way to go. For example if the peak is at 80Hz and you want to move it to 90Hz, the new capacitance would be smaller by the square of the frequency ratio, a factor of (80/90)^2, so 158u. You’d have to measure the caps too, since the tolerances are quite high.

Understood.

I'll have my Fluke 115 tomorrow.