Graphic Equalizer Display Filter - MSGEQ7

Thread Starter

gerases

Joined Oct 29, 2012
177
1) It depends upon whether you want a separate display for each channel or not.
No, I probably don't.

2) Jumpering as you indicated should work. What type of noise?
When the music is not playing, it sounds a bit like modem noise. Digital noise I would say.

Someone from Sparkfun told me:

As far as the modem-like noises, that seems like you are probably getting feedback introducted into the audio line with the addition of the second audio jack
.

Also what does this mean on the datasheet:

The multiplexer readout rate also controls the decay time (10% decay per read), so no external pins are needed for this function.
 

crutschow

Joined Mar 14, 2008
24,750
It's likely digital noise from the clock in the MSGEQ7 used for multiplexing and the digital filters.

You might try an RC low pass filter in the second output to see if you can reduce it. It also may help to take the second output directly from the source connector using a separate wire.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
It's likely digital noise from the clock in the MSGEQ7 used for multiplexing and the digital filters.

You might try an RC low pass filter in the second output to see if you can reduce it. It also may help to take the second output directly from the source connector using a separate wire.
Great, got it!


Sorry, but what does that quote from the datasheet mean, do you know?

The multiplexer readout rate also controls the decay time (10% decay per read), so no external pins are needed for this function.
 

crutschow

Joined Mar 14, 2008
24,750
Sorry, but what does that quote from the datasheet mean, do you know?
The multiplexer readout rate also controls the decay time (10% decay per read), so no external pins are needed for this function.
It controls the rate at which the readout changes with change in input. If you want to see the music beat, then you would set this to a fast decay time.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
It controls the rate at which the readout changes with change in input. If you want to see the music beat, then you would set this to a fast decay time.
So, I can control the decay somehow? Do you know how? What exactly is a readout?

Yes, I forgot there are single-chip audio circuits that perform those functions such as this, which should be fairly easy to incorporate.
I called Analog Devices about AN-934. It seems that it's not a chip but a circuit with several chips. The support person couldn't find AN-934 within the products they sell. His best guess was that it is probably an evaluation board of which they have limited quantities.
 

crutschow

Joined Mar 14, 2008
24,750
So, I can control the decay somehow? Do you know how? What exactly is a readout?



I called Analog Devices about AN-934. It seems that it's not a chip but a circuit with several chips. The support person couldn't find AN-934 within the products they sell. His best guess was that it is probably an evaluation board of which they have limited quantities.
The data sheet states that the readout rate is controlled by the frequency of the strobe input, which also then determines the decay rate. So a faster strobe gives a faster decay rate. The rate is limited to the values shown in the Strobe Timing Diagram.

The readout is what you are using to display the DC output from the chip.

The device referenced in the AN-934 app note reference I posted was the Analog Devices AD8336 variable gain amp. Did you look at the reference?
 
Last edited:

Thread Starter

gerases

Joined Oct 29, 2012
177
The data sheet states that the readout rate is controlled by the frequency of the strobe input, which also then determines the decay rate. So a faster strobe gives a faster decay rate. The rate is limited to the values shown in the Strobe Timing Diagram.
Aahhh. It's all coming together. Thank you!

The device referenced in the AN-934 app note reference I posted was the Analog Devices AD8336 variable gain amp. Did you look at the reference?
I see. I've found the AD8336. It seems to be surface-mount. That's a complication. Is there anything around that is through-hole?
 

crutschow

Joined Mar 14, 2008
24,750
...........................

I see. I've found the AD8336. It seems to be surface-mount. That's a complication. Is there anything around that is through-hole?
Not that I know of. Try doing a search for "low frequency AGC amplifier IC" to see if you can find something else.

This circuit is relatively simple and uses a JFET as a gain control element, which might also work for you.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
This circuit is relatively simple and uses a JFET as a gain control element, which might also work for you.
Great little circuit! It even has a description of how it works, which is rare. Thanks much! I just need to figure out how to make it 0 - 0.5V instead of 0 - 1.2V, which is probably as simple as lowering the value of C3?
 

crutschow

Joined Mar 14, 2008
24,750
Great little circuit! It even has a description of how it works, which is rare. Thanks much! I just need to figure out how to make it 0 - 0.5V instead of 0 - 1.2V, which is probably as simple as lowering the value of C3?
0 - 1.2V is the P1 pot adjustment range for the AGC output voltage. You just set the pot to give the 0.5V you want.

Changing C3 would just affect the frequency response of the control loop, which you don't want to do.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
The J176 they are using, according to the datasheet (found here), has a min of 1V and a maximum of 4V cutoff voltage. They are giving it 5V. Is that OK?
 

crutschow

Joined Mar 14, 2008
24,750
The J176 they are using, according to the datasheet (found here), has a min of 1V and a maximum of 4V cutoff voltage. They are giving it 5V. Is that OK?
If you look at the "Absolute Maximum Ratings" chart in the data sheet you will see that the maximum allowed Vgs is 30V, so 5V is fine. Any voltage above 4V simply maintains the transistor in cutoff.

Edit: I just noted they have a 10k ohm resistor (R7) from the op amp output to ground. This represents an output load and is not needed for proper circuit operation.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
If you look at the "Absolute Maximum Ratings" chart in the data sheet you will see that the maximum allowed Vgs is 30V, so 5V is fine. Any voltage above 4V simply maintains the transistor in cutoff.
I see, but why do they have under "OFF CHARACTERISTICS" those 2 numbers: min 1V, max 4v?

Edit: I just noted they have a 10k ohm resistor (R7) from the op amp output to ground. This represents an output load and is not needed for proper circuit operation.
Nice to know! Thanks!
 

crutschow

Joined Mar 14, 2008
24,750
I see, but why do they have under "OFF CHARACTERISTICS" those 2 numbers: min 1V, max 4v?
That's the range of voltages that will completely turn the transistor OFF. The spread is due to manufacturing tolerances from unit to unit. A particular transistor will turn OFF at a particular voltage somewhere between 1V to 4V. That value is unrelated to the maximum voltage that you can apply which will damage the part if you exceed it.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
That's the range of voltages that will completely turn the transistor OFF. The spread is due to manufacturing tolerances from unit to unit. A particular transistor will turn OFF at a particular voltage somewhere between 1V to 4V. That value is unrelated to the maximum voltage that you can apply which will damage the part if you exceed it.
I see, it will turn off between 1-4 and will take up to 30V.

Thanks!
 

crutschow

Joined Mar 14, 2008
24,750
While looking around for color organs I came across this project: http://arduino.cc/forum/index.php?topic=50021.0 According to the video, it works pretty nicely. But both on the schematic and the breadboard the inverting input (pin 2) is floating. Is that normal?

All the examples in the datasheet have it connected to ground.
It is not normal. The inverting pin input must be connected to something, likely ground in this case.
 

Thread Starter

gerases

Joined Oct 29, 2012
177
It is not normal. The inverting pin input must be connected to something, likely ground in this case.
My first error spotted online :)

I was looking at another color organ circuit by Forest Mims III in one of his books. The high pass filter looked like the file attached. Would you mind explaining to me in a few words how this filter works -- qualitatively. Please ignore the op-amp model -- it should be 1458.

I understand that when the frequency is over 1 / 2∏RC (where R=R12), the capacitor is close to a short (extreme case) and so the voltage on R12 is high and in this case we probably get an non-inverting amplifier with the gain of R11/R10. But what happens when the frequency is much below the rollover point?

Then the + input is 0V (extreme case) and it becomes an inverting amplifier? Doesn't make sense, I'm probably way off. Can you explain a few key points here?
 

Attachments

crutschow

Joined Mar 14, 2008
24,750
......................
I was looking at another color organ circuit by Forest Mims III in one of his books. The high pass filter looked like the file attached. Would you mind explaining to me in a few words how this filter works -- qualitatively. Please ignore the op-amp model -- it should be 1458.

I understand that when the frequency is over 1 / 2∏RC (where R=R12), the capacitor is close to a short (extreme case) and so the voltage on R12 is high and in this case we probably get an non-inverting amplifier with the gain of R11/R10. But what happens when the frequency is much below the rollover point?

Then the + input is 0V (extreme case) and it becomes an inverting amplifier? Doesn't make sense, I'm probably way off. Can you explain a few key points here?
You are making sense. It's just that the circuit is not a normal HP filter, instead it is a rather odd LP filter. It has a fixed gain at each of the frequency extremes with the LF gain higher than the HF gain. At DC or low frequencies the circuit is an inverter (phase = 180°) with a gain of 220k / 6.8k = 32.35. At high frequencies, well above the pole frequency, the amp becomes non-inverting (phase = 0°) with a gain of 1.

This circuit is commonly used as an all-pass filter (phase changes with frequency but gain is constant) if you set R10 equal to R11. For that configuration the low frequency gain is -1 (phase = 180°) and the high frequency gain is +1 (phase = 0).
 
Top