Hello,
I have a Divide down polyphonic organ that Ive modified by extracting the amplifier circuit, and taping in to the oscillator ICs directly. That signal is a strong line level, but I am wondering if it is better to install a coupling capacitor between the ICs and the the output.
Originally there was a .022uf capacitor before the extracted amplifier, and that cut out all the bass, so to be used with a small speaker inside the organ.
I tried a 2.2UF and that sounded full, but it seemed to cut out the very low frequency thump of the notes, which I really like. At the moment there is also a 1uf decoupling capacitor to ground, and a 1k resistor attenuating the audio signal.
It sounds better without the coupling capacitor but Id like to know if there is a better way about it. Without the coupling capacitor I can Measure DC voltage in the audio signal, but I can't hear any noise.
I could try a larger capacitor but I am concerned I could damage the now obsolete M082-AB1 ICs, so wanted to ask for some guidance before advancing.
I am using the organ connected to a DI box, but also sometimes with a VCF filter, with an estimated 100k input impedance, between the two.
Thanks
I have a Divide down polyphonic organ that Ive modified by extracting the amplifier circuit, and taping in to the oscillator ICs directly. That signal is a strong line level, but I am wondering if it is better to install a coupling capacitor between the ICs and the the output.
Originally there was a .022uf capacitor before the extracted amplifier, and that cut out all the bass, so to be used with a small speaker inside the organ.
I tried a 2.2UF and that sounded full, but it seemed to cut out the very low frequency thump of the notes, which I really like. At the moment there is also a 1uf decoupling capacitor to ground, and a 1k resistor attenuating the audio signal.
It sounds better without the coupling capacitor but Id like to know if there is a better way about it. Without the coupling capacitor I can Measure DC voltage in the audio signal, but I can't hear any noise.
I could try a larger capacitor but I am concerned I could damage the now obsolete M082-AB1 ICs, so wanted to ask for some guidance before advancing.
I am using the organ connected to a DI box, but also sometimes with a VCF filter, with an estimated 100k input impedance, between the two.
Thanks
It's early in the morning for me, but without a schematic, I cant imagine just where in this circuit "there is also a 1uf decoupling capacitor to ground, and a 1k resistor attenuating"...
Sometimes follow on devices provide their own DC blocking, because they dont know / cant be sure if any random upstream device did so. If you're confident about the 100K inputs on whatever you're connecting downstream, I cant imagine that blowing up anything audio driving it. Perhaps the thump you like is a feature of the DC you measure transitioning along with the notes.
FAIK.
You'll get a better answer if you can draw out a schematic, take a picture and post it here 😉
Taking significant current with a modification from an obsolete IC output that one doesn't have the specs for, and can only buy with a 90 lb wooden organ case attached, is a bad plan. Excessive current load can blow up IC's. Also excessive voltage swing from inductive loads (transformer DI units), can also.
If the ultimate load is truly 100kohms resistive, then the capacitance coupling the IC to it won't matter, .022 uf, 1 uf, 10000 uf, the current will be limited by the 100k input load. Say the load is an audio amplifier input which is usually >100k.
But if the load is 10 ohms, then 10000 uf to it could look like a 10 ohm load at 7 khz, and blow the divider or whatever IC you are taking the output from.
If you can't afford an amp to put the organ sound into, and don't have the spec of the "DI unit" input load, then the safe thing to do is put a current buffer between your organ output and the load. Say, a unity gain op amp circuit, to drive especially the capacitance of the cable you have coupling the organ and the whatever input. Inputs to op amp buffers are typically 10 uf or so, as the op amp IC's we use for audio, like 4558 or 5332, are not stable with DC inputs and tend to wander off in the DC realm. Audio is 20 hz or above, so the DC value of the input doesn't matter, that is why we use capacitors (cheap) or transformers (jensen DI units, expensive) to couple things. The op amp input will typically take small microamps or nanoamps, so they are negligible as loads in the current realm.
BTW just putting a 3 m long RCA cable as load on the organ takes significant current, as cable is about 100 pf a foot load or ~.001 uf for 3 m. So put any current driver board at the organ end of any cable, not at the PA amp end.
If the ultimate load is truly 100kohms resistive, then the capacitance coupling the IC to it won't matter, .022 uf, 1 uf, 10000 uf, the current will be limited by the 100k input load. Say the load is an audio amplifier input which is usually >100k.
But if the load is 10 ohms, then 10000 uf to it could look like a 10 ohm load at 7 khz, and blow the divider or whatever IC you are taking the output from.
If you can't afford an amp to put the organ sound into, and don't have the spec of the "DI unit" input load, then the safe thing to do is put a current buffer between your organ output and the load. Say, a unity gain op amp circuit, to drive especially the capacitance of the cable you have coupling the organ and the whatever input. Inputs to op amp buffers are typically 10 uf or so, as the op amp IC's we use for audio, like 4558 or 5332, are not stable with DC inputs and tend to wander off in the DC realm. Audio is 20 hz or above, so the DC value of the input doesn't matter, that is why we use capacitors (cheap) or transformers (jensen DI units, expensive) to couple things. The op amp input will typically take small microamps or nanoamps, so they are negligible as loads in the current realm.
BTW just putting a 3 m long RCA cable as load on the organ takes significant current, as cable is about 100 pf a foot load or ~.001 uf for 3 m. So put any current driver board at the organ end of any cable, not at the PA amp end.
Last edited:
Nooooo! 😱 😀
If you use a 22nF coupling cap and a 100k load, the time constant (RC) is 2.2 milliseconds. The rule of thumb is that, to preserve a square waveform going through a capacitor, the time constant should be ten times the period of the wave; so any square wave with a period longer than 0.22 mS will have its shape mangled by the RC coupling.
Translated to frequency instead of period, any square wave with a frequency below roughly 4.5 kHz will have its shape mangled. No wonder the OP hears a loss of "thump"!
The cure is simple, use coupling caps of 2.5 uF or bigger, to ensure square waves down to, say, 40 Hz come through unscathed.
For example, these 4.7uF ceramic caps are less than 50 cents apiece in Canada (they'll be cheaper in the USA), and will do the job nicely with some safety margin for frequencies below 40 Hz and/or loads a bit less than 100k: FK28X5R1A475KR000 TDK Corporation | Capacitors | DigiKey
Remember - this will only work as long as the 4.7 uF caps are feeding into a load not much less than 100k. If the load is much less, the "loss of thump" problem will return.
I agree with Indianajoe regarding the vulnerability of a rare and precious IC - do make sure to protect its outputs. I would suggest, say, a series resistance of 10k on the output pins (in series with the 4.7 uF cap), to protect it from overcurrent.
-Gnobuddy