U3, U5 are just DC-coupled buffers.
U4 mixes the DC part of the input signals while U2 mixes the AC part of the input signals.
The AC and DC are then mixed together; U1 is a buffer for the AC+DC mix.
Sorry Ron; Its part of the littleBits Synth Kit https://littlebits.cc/kits/synth-kit
The mixer mixes two channels together ether of witch can be audio or 0-5 VDC.
Almost all the modules use the same OpAmps to drive the outputs.
This is the only one I can see that has the voltage reference network.
U3, U5 are just DC-coupled buffers.
U4 mixes the DC part of the input signals while U2 mixes the AC part of the input signals.
The AC and DC are then mixed together; U1 is a buffer for the AC+DC mix.
U3, U5 are just DC-coupled buffers.
U4 mixes the DC part of the input signals while U2 mixes the AC part of the input signals.
The AC and DC are then mixed together; U1 is a buffer for the AC+DC mix.
The inputs and output are limited to +/- 5 volts by "ESD9B5.0".
>>The input can not respond to a negative signal. This gives me the idea it is not for audio. Will work for a voltage from 0 to 5V.
The inputs are divided down by pots. (10k and 10k pot) and referenced to ground.
>>Not good for audio.
The signal is split into two paths.
1) The signal is added together and low pass filtered. (DC) Then a gain of 2. Reference is ground again.
2) The signal is high pass filtered and added together. (AC) Then a gain of 4. Reference to 1/2 supply. Then another high pass filter.
Now the AC/3 and DC/2 are added together. (AC/3)+(DC/2) Why?
The output stage gives a gain of 2 and referenced to ground.
Every time it is referenced to ground it is not good for audio.
I should run this through spice to make sure........
The high pass and low pass filters are not of the same frequencies. Why?
Ignoring the loss of gain due to filter roll-off, from eye I get the output equal to 1.33*(AC1 + AC2) + (DC1 + DC2); perhaps the 1.33 is to make up for the reduction in gain caused by the HPF.
Ignoring the loss of gain due to filter roll-off, from eye I get the output equal to 1.33*(AC1 + AC2) + (DC1 + DC2); perhaps the 1.33 is to make up for the reduction in gain caused by the HPF.
The DC gain is set by R15 R22 and is 2x.
The AC gain is set by R10 R18 and is 4X.
Then when the AC and DC are added together:
DC:R13 R16 makes a 1/2.
AC:R14 and (R13//R16) will make a 1/3. (I think)
>The signal enters R14.
>R16 goes to ground.
>R13 goes to a AC ground. The output of the opamp is ground to a AC signal.
>R16 is in parallel to R13.
>The divider is 100k and 50k.
C10 0.01uF and 100k ohms R14 Why that frequency?
My time: it is a new day, (mid night) so good night.
For the AC (multipliers from left to right of circuit):
[LATEX](\frac{AC_1}{2}+\frac{AC_2}{2}) \times 4 \times \frac{1}{3} \times 2 = \frac{4}{3} \cdot (AC_1 + AC_2)[/LATEX]
For the DC (multipliers from left to right of circuit):
[LATEX](\frac{DC_1}{2}+\frac{DC_2}{2}) \times 2 \times \frac{1}{2} \times 2 = (DC_1 + DC_2)[/LATEX]
So the output is the sum of these two parts:
[LATEX]\frac{4}{3} \cdot (AC_1 + AC_2) + (DC_1 + DC_2)[/LATEX]
Guys this mixer is on the market and works. I just bought the kit for my son and he loves it.
It was designed in collaboration with Korg, one of the original synthesizer makers and it does work.
It may be an unconventional design and may not even sim right but you can try.
Its all open source hardware and here is a link to the files for this module; https://github.com/littlebitselectr.../INPUT/LB_BIT_i37_MIX/LB_BIT_i37_mix_v03(3_5)
Hope the link works.
I would like to make some of my own models some day to keep expanding what you can do with LittleBits, like meters and motor drivers and a longer sequencer. They are brand new but I think its a great idea.
That does not make it a good product.
Most of the noise makers for this line of products look like LM555 oscillators. 0 to 5V square wave out. This mixer will mix that 'OK' but it is not good for real audio. (sign wave)
Here you go.
It will not mix sign waves at ground. I hade to insert audio offset upward by 2.5 volts to get the amp to work. Other wise only the top half of the audio came through.
Here is the frequency response. The DC and AC are not the same gain by 8db. With that 1" speaker you can't hear 100hz so it does not matter. Could have been built with less parts.
Guys this mixer is on the market and works. I just bought the kit for my son and he loves it.
It was designed in collaboration with Korg, one of the original synthesizer makers and it does work.
It may be an unconventional design and may not even sim right but you can try.
It's not 'unconventional', it's simply appalling badly 'designed' (in fact I wouldn't use the term 'designed' for it at all) - if the kit works then it's likely to be a completely different circuit to the one you posted.
A mixer is an incredibly simple and easy thing to design and build, I find it hard to believe anyone with even minimal electronics knowledge could design one so poorly, there's so much wrong with it I don't have time to make a list
The littlebits website you posted the link to seems about as well designed as the circuit you posted, more a random series of unconnected pictures than anything else?.
Here is the frequency response. The DC and AC are not the same gain by 8db. With that 1" speaker you can't hear 100hz so it does not matter. Could have been built with less parts.
Hock up a audio source. (CD player) and listen.
Output of two sign wave sources at ground. 1khz and 3333hz
Output of the first buffer. Only 1/2 the signal is buffered. The other 1/2 is lost.
A mixer is an incredibly simple and easy thing to design and build, I find it hard to believe anyone with even minimal electronics knowledge could design one so poorly, there's so much wrong with it I don't have time to make a list
Wow I find it hard to believe its that poor of a design.
Can you point out an audio defect that I should be able to hear?
How hard would it be to improve it? and what advantages would the improvements provide?
if the kit works then it's likely to be a completely different circuit to the one you posted.
Its all on open PCBs, I could try and reverse engineer the boards with a good magnifying glass to check them but I have no reason to believe the schematics are not acerat.