| Author |
Message |
jumunius
Joined: Apr 19, 2010
Posts: 346
Location: San Francisco, CA
Audio files: 13
|
 Posted: Thu Nov 24, 2011 1:39 am Post subject:
|
  |
|
Hi Stew, thanks for the notes. I assume what you suggest will be more stable and efficient as a circuit, so would be worth a go. What I have now does work so far, although maybe I just don't know what I'm missing until I try this change.
| stewpye wrote: | | The section around the switch looks overly complicated and doesn't make sense. |
Hmm, well, it just provides 2 levels of hi hat mode, which I find useful to compensate for the various ways the EG might interface with other modules, the speed of the triggers it's working with, etc. Turning the cap on gives more distinction between notes, cap off is a little more fluid. Often the difference is subtle, sometimes it's more extreme.
Also, the more stupid reason I did it this way is it was the easiest to wire, and since the hi-hat mode was a late addition to my stripboard I was running out of room for extra connectors. Doing it this way, I just mounted the cap on my switch and connected it to the ground of other panel components.
Of course, perhaps redoing it as you suggest, and then changing R7 to a pot as I am considering might offer the most variety in the end.
| Quote: |
Here's what I'd do...
Bugger off the switch.
Put C3 between cathode of D2 and ground. Also connect anode of D2 directly to R9.
|
In theory, what's the difference between having C3 on one side of the diode or the other? If I move C3 to the other end of the diode, doesn't it then affect the closed hat envelope that is being generated by the same trigger signal, adding a very slight sustain? This may not necessarily be a bad thing, I'm just curious what the logic is.
I'll try to revise my schematic soon to make sure I'm reading your changes correctly.
_________________
-Jim |
|
|
Back to top
|
|
 |
jumunius
Joined: Apr 19, 2010
Posts: 346
Location: San Francisco, CA
Audio files: 13
|
|
|
Back to top
|
|
|
stewpye
Joined: Apr 30, 2009
Posts: 49
Location: Brisbane, Australia
|
| Posted: Thu Nov 24, 2011 1:17 pm Post subject:
|
|
|
Sorry, I made a mistake.
I meant to say connect cathode of D2 to R9. (which you have done), but R100 and C3 should be on the cathode too, so move R100 and C3 to the other side of D2 and that should be right.
Cheers,
Stew. |
|
|
Back to top
|
|
|
jumunius
Joined: Apr 19, 2010
Posts: 346
Location: San Francisco, CA
Audio files: 13
|
|
|
Back to top
|
|
|
stewpye
Joined: Apr 30, 2009
Posts: 49
Location: Brisbane, Australia
|
| Posted: Fri Nov 25, 2011 1:36 pm Post subject:
|
|
|
You'll need an input resistor to the op amp following the summing op amp, and the buffer before the glide/atttack pot is not necessary.
I really don't understand the purpose of this part of the circuit. If you really want an attack control you could build it into the envelope generator proper (on the other page of the schematic). The way this circuit is drawn it is not only an attack control is is also a decay control that will override your other decay controls. If you put a diode in parallel with P7 (pointing back in to the op amp) it will get closer to what I think you're trying to achieve.
Attenuverter circuit has a very high and variable output impedance. You could use a circuit that achieves the same thing but has the pot on the input of the op amp.
Stew. |
|
|
Back to top
|
|
|
jumunius
Joined: Apr 19, 2010
Posts: 346
Location: San Francisco, CA
Audio files: 13
|
| Posted: Fri Nov 25, 2011 5:40 pm Post subject:
|
|
|
| stewpye wrote: | | You'll need an input resistor to the op amp following the summing op amp, and the buffer before the glide/atttack pot is not necessary. |
D'oh! Forgot to add that back in. Thanks.
| Quote: |
I really don't understand the purpose of this part of the circuit. If you really want an attack control you could build it into the envelope generator proper (on the other page of the schematic).
|
Yeah, but I assume that would require three pots, whereas this way uses just one. I'm out of panel space as it is (according to my loose design principles that is). There would certainly be advantages to doing it that way, like having individual control. Then again, given that all 3 EGs can be combined into a single part (a la hi hats, shakers, animated cv pattern, etc) it could be nice to have one knob that simultaneously adjusts all 3.
I'll try your diode suggestion; that's easy enough to implement.
| Quote: |
Attenuverter circuit has a very high and variable output impedance. You could use a circuit that achieves the same thing but has the pot on the input of the op amp.
|
You mean simply swap P8 and R27? I guess I'd need an example of such a circuit as it's not really clear to me.
Thanks for all the continued advice!
_________________
-Jim |
|
|
Back to top
|
|
|
jumunius
Joined: Apr 19, 2010
Posts: 346
Location: San Francisco, CA
Audio files: 13
|
| Posted: Tue Nov 29, 2011 2:27 am Post subject:
|
|
|
I tried putting a diode parallel to the Glide knob. Theoretically it makes good sense to not want to lag the decays (as they can be set by their individual knobs) but I found that I prefer the non-diode approach when using the glide knob to sweep. You can hear a comparison in the 2 Glide test sound files below. Omitting the diode and allowing decays to lag means that the entire part washes out as the pot is turned CW (in this case only to 12:00). Adding the diode seems less musically appealing at least for the patterns I was trying. Arguably one could add the diode with a switch and pick their poison, and I may just do that.
In case anyone is curious, the signal flow for this file is:
Doepfer Digital Noise > CGS Steiner HPF > MOTM 190 VCA
EG CV to VCF and VCA
I also took a moment to record something more synthy, using the same pattern but a slightly different source and processing approach. See PercEG_PercussiveSynthTest.mp3 for this. There are 2 sweeps, and for the record neither use the diode described above. The first sweep uses the 47n pulse-extending cap, the second does. The difference is very subtle and obscured a bit by the addition of glide during the 2nd sweep, but suffice it to say switching off the cap slightly changed the groove.
MPS (Impact Voice) in non-trigger mode > CGS Steiner LPF > MOTM 190 VCA
EG CV to VCF and VCA
Lastly I tried adding the 100k resistor in parallel with the 47n pulse-extending cap. As Stew suggested, this changes the effect of the hi-hat mode, and fairly severely. I probably will go ahead and make this change but it will take a bit of testing to figure out how to adjust the other resistor values accordingly.
| Description: |
| Glide mode with addition of diode parallel to the glide pot |
|
Download (listen) |
| Filename: |
PercEG_Glide_Diode.mp3 |
| Filesize: |
546.73 KB |
| Downloaded: |
849 Time(s) |
| Description: |
| Glide mode without additional diode |
|
Download (listen) |
| Filename: |
PercEG_Glide_NoDiode.mp3 |
| Filesize: |
483.67 KB |
| Downloaded: |
799 Time(s) |
| Description: |
| EG test using a synthy sound source |
|
Download (listen) |
| Filename: |
PercEG_PercussiveSynthTest.mp3 |
| Filesize: |
1.34 MB |
| Downloaded: |
818 Time(s) |
_________________
-Jim |
|
|
Back to top
|
|
|
|
Forum index » DIY Hardware and Software
FAQ
Calendar
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
Chat Room
I'm not sure if it's more embarrassing to mess up the schematics or to have misbuilt the thing so severely in the first place. Certainly it's less work to correct the schematics, so I'm happy with the outcome.