SH-101 Envelope Sustain/Release Issue. Help!

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, lisa, Joe.

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Tue Sep 15, 2020 1:22 pm

so there is no legato or maybe portamento modes where the micro would tie notes together intentionally? what happens if you play a sequence really slow? have you tried loading in a new sequence? does gate high time vary with clock pulse width at all? or is it just rising edge sensitive? i feel like its either the microcontroller or a bad digital input. swapping out the microcontroller sounds like a reasonable option. other than that, the only other option is to go through all of the digital inputs and make sure they are doing the right thing, which sounds like a really tedious task as they are scanned matrices.
openmusiclabs.com

User avatar
makhho
Common Wiggler
Posts: 177
Joined: Mon Feb 05, 2018 11:53 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by makhho » Tue Sep 15, 2020 2:06 pm

Hmmm, that is an interesting thing you mentioned about portamento, since there are portamento modes in the sense that you can have ON portamento which functions from step to step and is changed in intensity / amount via the Portamento potentiometer on the bender board. You have AUTO which is portamento which follows LEGATO steps programmed into the sequencer (again intensity of legato controlled by the portamento potentiometer). And lastly you have OFF, which just turns if OFF and the portamento potentiometer has no effect.

Also regarding programming a sequence to play very slowly ( I actually think this is helpful since it really isn't doing what we expect here)

When I program a sequence with STEP RESTS (EX: C3, REST, C3, REST, C3, REST) and play the sequence slow using INTERNAL CLOCK to advance the seqeunce AND using GATE or GATE+TRIG, release does nothing to the ENV characteristics.

HOWEVER, when I play the sequence at roughly 1/3 of the way of the LFO/CLCK speed throw and higher, release effects the ENV characteristics and I can clearly hear it's effect on the VCA.
What does that mean?
This doesn't make any sense.

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Tue Sep 15, 2020 2:33 pm

how long is the gate usually? for example, if you program note, at what point in the note does the gate go off? so lets say you programmed C3, C3, C3... would the gate normally be a 50% duty cycle, or maybe 2/3 on, 1/3 off? right now its doing almost always on, with only 5ms off. when you put in a rest, im assuming that is supposed to be 100% off for that step.

so at slow rates, the release doesnt activate, but as you turn the LFO/CLK up, then release starts to activate? can you probe the gate out and see what changes across that range.

im wondering if ive missed something in the ENV section. ill go back and have a look.
openmusiclabs.com

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Tue Sep 15, 2020 2:36 pm

also, what is the duty cycle on the clock signal to the microcontroller? does that change with LFO frequency?
openmusiclabs.com

User avatar
makhho
Common Wiggler
Posts: 177
Joined: Mon Feb 05, 2018 11:53 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by makhho » Tue Sep 15, 2020 2:38 pm

Guest -- I'm headed to a recording session across town and won't be able to check on things you asked about until tomorrow morning. I will get back to you promptly then and with more details about all of this.

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Tue Sep 15, 2020 3:09 pm

sounds good. ill have more questions by then, too. i think i see a way that it could be a problem with the ENV.
openmusiclabs.com

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Wed Sep 16, 2020 12:20 am

so there is some possibility that either the clock signal or TR32 isnt doing what it should. take a look at the signals on the ENV source selector switch (or R147 which it connects to). see if it switches between nothing, the trigger, and the clock signal. is the clock signal 50% here? then look at the signal on the collector of TR32, does it follow the clock signal (inverted)?
openmusiclabs.com

User avatar
makhho
Common Wiggler
Posts: 177
Joined: Mon Feb 05, 2018 11:53 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by makhho » Wed Sep 23, 2020 6:11 pm

Sorry for the late reply on this, guest -- finally found some time to dig deeper into the 101 troubles.

For starters, I discovered a totally unrelated issue that I didn't catch somehow after the whole episode with the reverse polarity walwart. I usually don't have any of my gear at maximum throw of the volume knob, but it turns out that the VOLUME pot was not responding and was cutting out at the output jack when the volume pot was at a throw of 3/4 and higher.

I originally thought it was a bad volume knob, but I swapped it out with a working NOS replacement that made no change to the output behavior. After that I looked at the M5218L on the bender board, but that wasn't the issue. Next thing in the signal path from the Bender board to main board is the BA662 and TR28 (which I already replaced). Issue turned out to be a bad BA662 in the VCA section, which I confirmed by replacing it with a CLOWN ;^).

-------------------------------------------------------------
SO, now on to addressing your questions about gate signals at PIN 38 and R147.

The duty cycle of the gate out signal when running the sequencer with various sequencing considerations is as follows:

1) When programming a C3,REST,C3,REST,... style sequence, PIN 38 drops from +5VDC to 0 for the duration of the sequenced C3 note (ONE FULL CYCLE OF ON/OFF illumination of the LFO LED), it then returns back to +5VDC when the REST "note" is played and stays at +5VDC (ONE FULL CYCLE OF ON/OFF LFO LED ILLUMINATION) until C3 is sequenced, etc.

PIN 38 shows a +5VDC constant gate signal for the duration of a REST step.

2) When programming a C3, C3, C3,... pure note style sequence, PIN 38 is at +5VDC and then drops to 0VDC when the first note of the sequence is played and stays at 0VDC for the full duration of the ON/OFF LFO LED ILLUMINATION CYCLE. At the beginning of each successive step in the sequence, PIN 38 shows a +5VDC pulse signal. After it drops back to 0VDC it stays there until the start of next step, where again it does this +5VDC 5ms pulse.

With straight note sequencing like in 2), the LFO ENV control setting is showing the same gate behavior at PIN 38, however again, I can modulate the VCA with a proper ADSR, regardless of LFO clocking speed when switched to LFO ENV.

Regarding R147 and PIN38 --
3) With a C3,C3,C3... sequence, PIN 38 shows the gate signal behavior of case #2 (as expected), probing R147 shows an inverted 0VDC pulse for 50% duty cycle (50% of the LFO LED ON/OFF cycle) with each triggered note. The signal at R147 goes back up to +5VDC when the LED is "OFF" or the other 50% of the duty cycle.

Kinda looks like this on two probe channels, i|____|----i|____|----i|____|----i ....., where "i" is the +5VDC 5ms pulse signal at the beginning of each triggered step coming out of PIN 38, and |___|----- is the 50% duty ENV pulse cycle seen at R147.

4) Regarding the ENV source selector switch -- R147 shows 0VDC when on GATE, +0.5VDC on GATE+TRIG, and when in LFO follows the LFO clock with like a +4VDC pulse at 50% duty cycle.
Individual key presses no matter how quick or slow show 0VDC at R147, but when pressing lets say adjacent keys quickly, R147 shows +8.8VDC signal.

5) Yes, the collector of TR32 follows the clock signal inverted when a sequence is played. TR32 shows nothing when a sequence isn't being played, but we wouldn't expect it to show anything, right? When in LFO ENV control mode, R147 shows the clock signal running at all times regardless of a sequence playing or not.

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Thu Sep 24, 2020 2:24 am

as best i can tell, that all seems right. can you take a look at page 14 here and let me know what you think:

http://www.synthfool.com/docs/Roland/SH ... anuals.pdf

it shows that the release does not become active unless youre in LFO mode.
openmusiclabs.com

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Thu Sep 24, 2020 2:42 am

theres another diagram on page 17 showing similar operation (pdf page numbers). it does show a slightly wider reset pulse on the gate, but that might just be for legibility in the diagram. its assuredly not 50/50 or even 75/25 in the picture.
openmusiclabs.com

User avatar
makhho
Common Wiggler
Posts: 177
Joined: Mon Feb 05, 2018 11:53 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by makhho » Thu Sep 24, 2020 2:38 pm

I think I am either confused, dense, or both.

If I am understanding this correctly, according to the owners manual, ADS are the only ENV parameters that are worked through while pin 38 GATE is high. Release doesn't initiate until the gate is low. If this is true, then what I am witnessing coming off of PIN 38 on my 101 is not right? As I'm just seeing a 5ms pulse on PIN 38 followed by a high gate until the next step is triggered with a 5ms pulse present. There is a basically a 100% duty cycle of the gate being high between each sequenced step? But the manual is saying that there is a point where the gate drops low and the release is initiated on page 14 (unless this is depicting keyboard playing/LFO playing/or GATE IN triggering?).
I feel like this doesn't happen to me when playing a sequenced pattern internally. There is no point in the time from sequenced step to step that the release has any contribution, in other words the gate never goes low at any point in the time between step to step.

Is the behavior I described actually what is detailed in the owners manual? When compared to the owner's manual gate/timing diagrams -- am I confusing and misinterpreting what is coming off of PIN 38 GATE OUT from the micro controller and what the ENV gate signal looks like at R147?

Also, page 17 shows that there is a minor minor contribution from the release to the ENV when in GATE+TRIG control mode between each sequenced step. It is correct in depicting that release has a contribution when the rest occurs.

Is it possible that I am all wrong on this? If this is the case, then Guest, it looks like you were right back when you asked me if this ADSR behavior in Gate+Trig/GATE ENV control mode was just a feature of the SH101. The owners manual appears to spell it out clearly, but it does not clearly describe how the ENV control modes respond to sequencing via the EXT CLK IN, but I checked and it is ostensibly identical to sequencing via the internal clock, which the manual covers.

If this is all correct behavior of the 101, then the only explanation I can provide for this paranoia run-around is that I've used the instrument for so long that I just got used to the fact that the ADS were the only parameters of the ENV in GATE+TRIG & GATE control modes that were contouring the VCA/VCF and it was just second nature to adjust or not worry about the release when using it.
After the reverse polarity incident I must have been hyper-paranoid of functionality changes, so much so that it made me consider the UI from a "different" perspective, making me want to look for issues that were non-existent.

Strange.

Since reading the owners manual, I asked two friends with 101s about this ADS vs ADSR ENV feature. We compared 101 behavior over the phone, and they too were absolutely surprised about this fact; seems that they also glossed over this. Which is strange because they both own modular gear and have various ENV/transient generators/contouring module options and should have registered the behavior.

The fact that the LFO/External gate IN/Keyboard playing are the only ways to make the all of the parameters of the ENV modulate the VCA and VCF is strange, as this is something I'd personally want across all of the ENV switch control settings.

I feel dumb, though I'll admit that I'm relieved; my sincerest apologies for having made you come on this journey with me.

User avatar
guest
Super Deluxe Wiggler
Posts: 4512
Joined: Mon Aug 19, 2013 11:49 am

Re: SH-101 Envelope Sustain/Release Issue. Help!

Post by guest » Thu Sep 24, 2020 3:07 pm

yeah, i think this may just be how it operates. ill agree that it is a bit strange, it would be nice if it was switchable. i think you can sort of make it switchable by putting a rest between each note in a sequence, or just use the LFO mode (maybe with external clock). if you wanted to mod it, it wouldnt be that hard to put in another control that waited for a fixed amount of time after a note-on, and then shut the envelope off. this fixed time could be set with a pot.
openmusiclabs.com

Post Reply

Return to “Music Tech DIY”