That’s because it’s a ‘true bypass’ pedal. Your pickups are probably around 6k so 10k may actually be worse.
That doesn’t make any sense. (though I doubt that’s relevant now anyway)
If its a ‘true bypass’ pedal (as it, and you, claim it is) then in bypass I would expect the guitar is just connected straight to the output jack. The pedal electronics are not in the circuit at all (so the pedal circuitry I/O impedance, or ‘impedence’ ) should not matter. In bypass I would expect it to sound muddy, just like going straight into the ADC
With the comp off, it was super muffled, just like going directly into the mic input.
Yes, I agree. What I meant was ‘in bypass it sounds muffled because it’s true bypass’ and ‘when it’s on there’s not much difference because it’s still 10k’
‘when it’s on there’s not much difference because it’s still 10k’
The pedal’s input impedance e.g. that seen by the guitar, is 500K not 10K (so I would expect that to sound better than going direct to the ADC. Its not really possible to speculate about any other effect that the pedal may have (e.g. due to driving the ADC input), without knowing more about the circuit used).
Sounds as though if hifiberry can make the required changes to the ADC circuit then everything should be fine with a guitar input - I might even be tempted to buy a DAC+ ADC myself if they make it a standard option - I’ve been very impressed with their original DAC+
Using this model :
it’s possible to get some results. Assuming the input is around 10k (nominal line input impedance) Zs = Zl.
Vl = Zl/(Zs + Zl) * Vs
Vl = 1/2 * Vs
or the signal has been halved because the output impedance is too high for this input. The guideline is to make the input impedance ten times the value of the preceding output stage.
In the case of an electric guitar pickup there is an inductor in series with Zs making a low pass filter. Using a much higher value for Zl, like the 1 Meg mentioned above puts the filter cutoff higher, preserving treble.
I’m assuming you mean “The input of the ADC circuit is around 10k” (we don’t actually know that, unless it’s in the DAC+ ADC spec and I missed it, but it is likely lower than e.g. a guitar amp).
The stated input impedance of the pedal, by which I mean the comp-engine compressor pedal, is 500K, which, for the reasons you mention regarding the pickup, would therefore be expected to have less of an impact on the guitar’s tone than connecting the guitar direct to the ADC input. I think we’ve established that is likely what is happening. If hifiberry can make a version with a more guitar-friendly input then it will probably work just fine.
I’m doing some further experimentation with various configurations and figured “Well if the hifiberry mic input is muffling the higher mids and above, then what if I just use a counteracting a-EQ instance to crank up those frequencies?” And the result is actually pretty close to normal.
Let me frame my question this way (and I don’t know if I’m asking it correctly):
Is using the mic input lossy (information is lost by the mic input) or just biased (information is not lost, it’s just shaped strangely due to the impedence issue)?
If it’s the former, then that’s bad. No amount of post processing can get it back. If it’s the latter, then that’s good and using EQ after-the-fact can fix it.
Because it is shaped strangely due to the impedance issue, it is in some form lost. Even making up with EQ will get you close, but never completely back, and depending on the EQ you may introduce phase issues. Always better to start with the best source as a result.
Sounds like more of the latter (just biased, not lossy). I understand what you’re saying that once it has been shaped, it’s going to be essentially impossible to recover the original signal. But for my purposes, because either none or a negligible amount of signal is truly lost, that’s good enough. I can use EQ to get it back.
And hopefully in the near future I’ll have a hifiberry board with a real guitar input.
The pickup’s impedance is frequency dependent, but that relationship is rather complex and the guitar coil and low-impedance input form a 1st order resonant low-pass filter. That is definitely lossy. The signal is attenuated, low-pass filtered, and parts of it will vanish in the noise-floor. Also some sound characteristics change. For example the sustain is shortened when the coil pickups are not feeding a Hi-Z input.
If you heavily process the signal afterwards (loads of compression, distortion, chorus, etc.), you might be able to get away with undoing the effect of the low-pass. But keep in mind that this also depends on the pickup type and will be different for different guitars.
Meanwhile just inserting a 470kOhm resistor before the input, terminated with a 22kOhm to GND and ramping up the input gain by ~27dB will likely produce a better sound, compared to toying eith EQs.
First off, it is crazy how much of a brainiac you are.
Second. I will do this, but oh man, I am not an electronics guy. What’s the ELI5 version? Take a guitar cord, cut it, strip the wires, connect ? to ? and ? to ?. What component increases gain?
Third, do you think there is a pedal that does all this already? If so, I’m happy to buy it as a stopgap until my custom hifiberry is made.
The phrase you want to look for is “buffer pedal.”
I did a quick search for guitar buffer pedal and found this one for about the price of a NYC hamburger:
I can’t vouch for it, I really mean just found it seconds ago, but that is the kind of thing to look for. For $17 and free shipping I would give it a try.
Well I have a DI box, but it is USB and I’m attempting to create real-world conditions by using this hifiberry. So I’d rather put something in the signal chain from the guitar to the hifiberry instead of swapping input/output entirely.
Does the wattage rating on these resistors matter? I’m seeing 1/4 w, 1/2 w, etc on amazon.
So in that case I would get a real DI box, this is standard on the stage. Anyone playing live and plugging direct into a sound system should have one, and some of the cheaper ones aren’t to expensive. Something like this:
Connected to the header marked ‘5’ in your image above, with the gain jumpers set to +32dB should be pretty close.
Seablade
EDIT: I should clarify, a USB DI can still be a DI box, the term has become far more generic than the original devices that just took a high impedance unbalanced signal and output a low impedance balanced signal, but obviously it won’t help in your case unless you can run the USB DI into the Raspberry PI directly (I assume you have already tried this?)
Much appreciated. How would I go about connecting this to the “5” from my image? Is that connector on the board some sort of standard connection? If so, what’s it called?
And why not just go from 1/4 to 1/8 into the mic jack? Is that problematic for some reason?
I would STRONGLY recommend not touching Behringer DIs with a well insulated pole.
While Behringer has released some usable products recently, their DIs were so notably bad in the past that I would not even use them for skeet shooting.
The DI I linked to is still considered one of the ‘cheaper’ DIs but is a workhorse you will find on most stages.
) should not matter. In bypass I would expect it to sound muddy, just like going straight into the ADC