808 hihats need some help

Hey, I have started working on some 808 hihats. I simulated some stuff and the envelope is already working, I think :slight_smile:

This is the envelope for the open hihat, which is affected by the closed one. I tried the same as in the hexinverter mutant hihat: adding a switch which determines if this happens and to what extent.
It was only a question of finding the right resistor :wink: So:
red line:
no effect of the closed hat (the closed hat circuit is disconnected from the open hat, simulated)
green line:
small effect of closed hat (closed hat circuit connected via 300k), this should dampen the sound of the open hat somewhat when the closed hat is triggered
blue line:
big effect of closed hat (closed hat circuit connected via 10k), this should completely silent the open hat as soon as the closed hat is triggered

Maybe I need to changed some values, but the principle works :slight_smile:

Now I need to decide if there is any point in replacing the so called swing-type-VCAs of the original 808 with “proper” (OTA-based) VCAs, as has been done in the hexinverter. What benefit would that give? Do the swing-type-VCAs not close completely or something like that?

Also, I want to filter the noise with a bandpass, but I have not yet found a good solution… I want something simple, no need for CV, just a simple knob should do, but I have not found a good filter design. I think I need to go with a gyrator based filter, because I would prefer not to use a dual gang pot and also no OTA…

But I had no success in designing it…
Here is an FFT of the noise:


It looks like it’s mostly around 4 to 10 kHz, so I want the filter to be in that range too.

When I use the standard tunable gyrator bandpass from Elliott Sound Products I get this:

I would like to shift that a bit higher! But when I reduce C4 or R1, the overall gain(?) drops:

If I lower C1, the bandwidth gets very high and I think I don’t want that:

So, I am clearly missing something, but I don’t really know what! Any ideas? :slight_smile:

The bandpass filter from the YuSynth site I posted earlier

is pretty simple, if you use the single stage because you don’t need 12 dB/oct. And the equations should give you what you need to figure out component values to achieve the frequency and bandwidth you want.

1 Like

Yes, but with that I can only change the Center frequency of the band when changing R1 and R3 at the same time using a dual gang, right?

As I understand it, this multiple feedback filter works as a slightly more complicated inverting amplifier:

Let’s ignore C2 and R3, then we have a simple inverting amp with gain determined by C1 / R1. At high frequencies C1 gets effectively smaller, so the gain drops.

image

Now ignore R1 and C1: same inverting Opamp, but now with resistor and cap switched. Gain is now R3 / C2. For low frequencies C2 “opens up” (high resistance). So here the gain drops as well.

image

So, R3 and C2 determine the lowpass of the whole filter while R1 and C1 control the high pass. To move both through the frequency range I need a dual gang pot. The fixed filter bank is called fixed for a reason :wink:

That’s why I have been looking at gyrator filters :slight_smile:

Ah, hadn’t realized you wanted to vary the frequency. Yes, I think you’d need dual gang to do that. Nice explanation of the circuit, by the way.

2 Likes

I think I found what I wanted :slight_smile: I just played around with the original 808 schematic and there is a filter directly after the hexinverter chip that makes all the square waves for the “metallic noise”. Turns out that if you replace the 560 ohms to ground with a pot, you get quite a nice resonant bandpass there! :slight_smile:

3 Likes

Nice work - looks like a logarithmic pot would work well here based on the numbers you simulated.

3 Likes

Progress is slow, but I have an idea for the panel now:

5 Likes

Are those capacitive tap pads, or just fancy graphics?

1 Like

This comment might just throw development back another 3 to 14 months :joy:

It’s just graphics, the pure FR4 should be exposed and an LED behind that :wink:

How would one go about an analogue touch button, btw? XD

3 Likes

Nice ! Im curious to have more details of your 808 Hi-Hat version, your various functions are looking really great !!

I have build mine with Eric Archer’s schematic. Sound really good !

http://www.ericarcher.net/wp-content/uploads/2014/07/tr-808-hihat-diy-project-revised-dec-2009.pdf

1 Like

I’m not sure to understanding your request but maybe something like that ?!

I was hoping you would be telling me!

2 Likes

I was thinking about a (capacitive) touch plate, like in Sam’s latest sequencer version :wink: I have never looked into that, he does it with Arduino code, but there should be a way to do it without any microcontroller, I think. But not on my module this time :slight_smile:
With your push button, I guess you might have an issue when there is a jack connected that pulls the input to ground (because it’s not triggering) then the push button has no effect! I have this problem on an envelope generator I made. I think it needs a transistor at least…

3 Likes

I am also working of that schematic! But I added adjustable decay length for the closed hat and I made the ‘exclusive’ function (where the closed hat mutes the open hat) optional, so you can enable it with a switch. Oh, and I added the tone knob, which affects the bandpass filter for the noise. I am still working on the output level, because I would like to be able to add a bit distortion at the end, but I am not sure how I want to do that… diodes in the feedback? diodes to GND? no diodes, just let the opamp clip?

I am also wondering how to best normal the accent input to 10V (or maybe 12V?). Just use the switch from the jacksocket? do I need to add a resistor to protect stuff that I put into that? A simple pullup won’t work, because there is a voltage divider after that…

4 Likes

A voltage through a protecting resistor to a jack tip switch is a simple pullup. But presumably the voltage divider resistance (on the top) is some tens of kΩ, right? You don’t want to connect a rail directly to the tip switch because it’s likely to get shorted to ground when you plug cables, but it wouldn’t have to be huge, maybe 470R or even less would be fine.

2 Likes

This looks pretty exciting.

I am still looking for a good analog hi-hat schematic so I’m curious if you are willing to share yours. I don’t want to settle for the “white noise bursts” and instead mix some pulse waves for a metallic edge.

Do you have some audio samples of your module already?

2 Likes

It’s not done at all xD I have only done a little breadboarding, but only the noise and filter part. The whole circuit is too much for me to breadboard. I am doing simulations in LTspice. I will of course share what I got when it’s a bit more progressed :slight_smile:

3 Likes

I guess you mean a jack where a 5 V trigger goes in, but it’s normalled to ground so if there’s no cable plugged in it’s grounded. If you can change that to connect to ground through a resistor then a push button can override it. In fact if there were no pulldown resistor you’d need to add one to @saint_et_moudulard’s diagram since the button doesn’t ground the input when it’s open.

3 Likes

This is the relevant circuit section:

Screenshot from 2021-11-29 14-29-38

And you mean something like this could work?

1 Like

This is from the envelope that I built. The push button works, but only when there is no jack in the socket! As soon as I connect a jack there, the push button does nothing. Maybe it would be enough to put the another resistor between the gate in (jacksocket) and the part where the button connects. I think what happens is, that the thing on the other end of the cable (beatstep pro, midi2gate,…) outputs 0V (GND) when it’s not triggering, and therefore the pushbutton can not get the voltage up, because there is no resistor…

1 Like