Thanks! I'll put in some research in tonight.
Actually I'm not so sure. Usually these capacitors are there to remove the DC offset of the signal, the positive bias, but with higher values (1uF or 10uF).
And as you mention that at 60Hz you get a weak signal, I tough that smaller values of C1 could start to attenuate these low frequencies also.
10nF has a reactance of about 265K Ohm at 60Hz, what is considerable. For higher frequencies the reactance is much lower (Capacitive Reactance Calculator).
But I see that they recommend that value (10nF) in the datasheet of the MSGEQ7, so I don't know. You can just try.
The msgeq7 datasheet indicates that the typical clock rate value on CKIN is 165 kHz, with 33pF/200k.
However, it indicates two more values: minimum = 145 kHz and maximum = 180 kHz
For the typical value of 165 kHz in CKIN, the bands are 63 Hz, 160 Hz, 400 Hz, 1 kHz, 2.5 kHz, 6.25 kHz and 16 kHz
With that basis, you can play with the values of C and R connected to CKIN.
In the most recent review of the sparkfun msgeq7 shield, there is the correct circuit, I replicated it with SMD resistors and capacitors and it worked the first time, especially there is a response in the 63 Hz band.
1 Like
I'll have to look into Capacitive Reactance. I'm not familiar with it. I like that this project is giving me a lot of practical things to research. Thank you for your suggestions and advice! I wasn't able to really work on it last night because my capacitors didn't come in yet.
Read the answer of @TFTLCDCyg.
The freq on CKIN pin is a very important to determine input frequences correctly. There are a few projects, where changing the CKIN frequency is used to shift the MSGEQ7 frequency characteristics and create a spectrum analyzer of more than 7 bands.
Accordingly, if you observe incorrect band frequencies in your project, you need to adjust the CKIN frequency.
(It is what i meant in my #11 message, but accidently mixed up CKIN and STROBE inputs.)
Yes. Also in the linked document the circuit seems to be the correct one, or more up to date, some things are different. And also, C1 (C6 there) has grown to 100nF.
Some time ago I explored the possibility of adding a device that could generate two different frequencies, to feed the CKIN of two msgeq7, within the frequency tolerance range. However, my knowledge of electronics was (and still is hehehe) somewhat limited. So I was lost on the correct way to connect such a device.
The Si5351 breakout seems to be a good candidate.
It is possible to generate one frequency with the breakout and the other to keep it with 33p/200k in one of the msgeq7 chips.
In this way it would be feasible to extend the audio spectrum, from 7 to 14 bands, for the same audio input. However, it's still necessary to explore the resulting frequencies of the msgeq7 attached to the Si5351.
It is a quite possible, see the link below for a project that uses a Si5351A board to shift the frequency of the filter bands of two MSGEQ7:
Thanks for the reply! I haven't had much time to work on this recently and I haven't received my oscilloscope yet. I should be getting the scope today and might be able to do more testing.
In my more recent replies in this thread I might have been too fixated on the MSGEQ7. That datasheet is where I based my original circuit on from my first post. There are some obvious mistakes in the schematic I posted but the intent was to replicate the circuit you linked to. The clear difference being that I was using a MAX9814 as my input signal vs what seems to be an output from a 3.5mm jack. My first post has a video of the results from testing with a 200k and 33pF on the CLK pin. I had 3 MSGEQ7s and they all acted in the same fashion. I may have a bad batch of MSGEQ7s but they do work at 175k and 33pF on the CLK pin. You can see my testing in this post. In this video, you can see how my 63Hz signal does work when connected directly to the computer's headphone jack:
I think my issue is the interface of the MAX9814 and the MSGEQ7. That is also why I'm waiting for my scope because as @gromit1 suggested:
I had an obvious issue with my voltage divider originally that @gilshultz helped me fix. I'm kind of leaning back towards something being wrong with my voltage divider if I can confirm that the MAX9814 can pick up the 63Hz frequency.
I answered originally based on what you said but this lead me to start looking at the CLK pin. So your intent worked. Thank you!
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.