EURO x0x-heart - Complete "Kit"

[guest]

If I set the x0xheart to C2 (VCF output set to INT) and the Dixie to C2 and then apply a constant 5v to the Accent Gate input with a repeating gate from the x0xb0x to the x0xheart’s gate input (the pitch rise only happens when I send gates to the x0xheart’s gate input) the pitch of the x0xheart rises up and is never in tune with the Dixie. But it’s closest in pitch with the Dixie at the end of each gate.

this is a really good test here. so both the x0xheart and dixie shift up in pitch, but the dixie comes down faster? maybe the x0xheart doesnt completely come down? can you post a picture of your x0xheart (both sides)? also, with your scope tests, what was your V/div settings? if you can set your scope to DC couple, 50mV/div, and then adjust the DC offset so that its at -2V, you should be able to bet a clear view of what the slid CV out is doing. you could put one probe on each side of the 1k, and then youd see how well they track. some scopes limit the DC offset value based on the V/div setting. so if you cant get the -2V offset at 50mV/div, try 100mV/div. again, were expecting 100mV movement at the most.

its sounding like there is a bit of extra current flowing into the CV pathway when accents are active, as a result of the envelope circuit. U12 could be the culprit, as it connects to both the accent circuit and the CV input circuit. if there is extra flux in this area, or that switch has internal leakage, then that could cause a shift when accent is engaged.

[morbank]

If I set the x0xheart to C2 (VCF output set to INT) and the Dixie to C2 and then apply a constant 5v to the Accent Gate input with a repeating gate from the x0xb0x to the x0xheart’s gate input (the pitch rise only happens when I send gates to the x0xheart’s gate input) the pitch of the x0xheart rises up and is never in tune with the Dixie. But it’s closest in pitch with the Dixie at the end of each gate.

When I said “ the pitch of the x0xheart rises up and is never in tune with the Dixie. But it’s closest in pitch with the Dixie at the end of each gate” I misspoke... I should have written...

In this test when a gate is sent to the x0xheart’s Gate input the pitch of the x0xheart (VCF output) stays at C2 (no pitch shift).

When a gate is sent to the x0xheart’s Gate input the pitch of the Dixie’s output (fed by the Slid CV output) starts high with each gate and then the pitch falls to C2.

When I had said they are never in tune with each other that was because I was using gates from the x0xb0x and the tempo of the gate sequence wasn’t slow enough so the pitch didn’t have enough time to fall to C2. But if I manually trigger a gate the pitch of the Dixie’s output does eventually fall to C2.

Apologies for the confusion.

I’m taking some photos of the boards now and will attempt to get those scope measurements as well, but I just wanted to correct my mistake before we moved forward.

Also, in case it’s pertinent, the x0xheart kit that I bought came with the SMD x0xheart board pre assembled. I looked at the soldering on U12 but it looked solid to me.

[guest]

excellent, thanks for the update.

[guest]

also, a picture of how the slid CV out jack is wired up and connected would be great as well. does the time it takes for the dixie to match the x0xheart vary with the decay knob setting? since the x0xheart itself is keeping tune, im thinking it might not be an issue with the x0xheart, per se, but rather related to how much power the x0xheart is drawing. its all very odd. also, if at any point you want to just bypass the 1k and call it day, thats a fair response.

[morbank]

also, a picture of how the slid CV out jack is wired up and connected would be great as well.

Will do. I won’t be home most of today so I’ll have to post these tomorrow.

does the time it takes for the dixie to match the x0xheart vary with the decay knob setting?

Yes, it does!?!

since the x0xheart itself is keeping tune, im thinking it might not be an issue with the x0xheart, per se, but rather related to how much power the x0xheart is drawing. its all very odd.

It really is!

also, if at any point you want to just bypass the 1k and call it day, thats a fair response.

Ha! Yeah, I was thinking that if you don’t see anything in these upcoming pics that it might be time to move on. Regardless, you have been incredibly patient and helpful with this issue and I super appreciate it!

[morbank]

this is a really good test here. so both the x0xheart and dixie shift up in pitch, but the dixie comes down faster? maybe the x0xheart doesnt completely come down? can you post a picture of your x0xheart (both sides)?

X0xheart by Choking Hazards, on Flickr

X0xheart by Choking Hazards, on Flickr

X0xheart by Choking Hazards, on Flickr

X0xheart by Choking Hazards, on Flickr

X0xheart by Choking Hazards, on Flickr

also, with your scope tests, what was your V/div settings?

I was just using my Mordax Data at the 1v/div setting. Anything lower and I couldn't see the trace (off screen).

if you can set your scope to DC couple, 50mV/div, and then adjust the DC offset so that its at -2V, you should be able to bet a clear view of what the slid CV out is doing.

I don't believe the Mordax Data has DC offset. I also have an old Tek 465B scope, but I don't that it has DC offset either :(

you could put one probe on each side of the 1k, and then youd see how well they track. some scopes limit the DC offset value based on the V/div setting. so if you cant get the -2V offset at 50mV/div, try 100mV/div. again, were expecting 100mV movement at the most.

The lowest I can go on my 465B is .2v/div. In this video the top trace is the 'post 1k' channel and the bottom trace is the 'pre 1k' channel...
https://www.flickr.com/photos/choking_h ... res/0pyhL4

[guest]

that scope shot is super convincing, although i have no idea what could be causing that. furthermore, what would cause a voltage to go up on the other side of a resistor??? that would mean current is being injected into the opamp from an outside source. was this just going to the scope, or was something else connected to the slid CV out? was this with the module all bolted together as in the last shot above? is it possible the jack is touching something on the pacemaker? its really tight in there.

[morbank]

that scope shot is super convincing, although i have no idea what could be causing that. furthermore, what would cause a voltage to go up on the other side of a resistor??? that would mean current is being injected into the opamp from an outside source. was this just going to the scope, or was something else connected to the slid CV out?

Just to the scope.

was this with the module all bolted together as in the last shot above?

Yes.

is it possible the jack is touching something on the pacemaker? its really tight in there.

It is possible. I put some electric tape down underneath the jack and then took it apart to make sure nothing was pushing through the tape. It does fit, but just barely. I’ll take another look to make sure...

[guest]

if you remove the faceplate, so its not touching anything, and test again, does it do the same thing? there are only 2 ways the signal could be going up: 1. a signal coupling in on the jack, or the wire to the jack. this could be capacitive or resistive. if it was resistive, it would be on the order of 100k. the odd thing is, this is the VCF envelope signal, and that only exists on the ACCENT, ENVMOD, and ACC->RES knobs on the pacemaker. if the signal was coupling through on the x0xheart itself, the 1k wouldnt matter and youd see it right at the output. 2. the ground connection to the jack isnt good, and some current flow boosts the ground up somehow. but, you previously measured the ground connection at the jack to be less than 1ohm, so this is not likely.

[guest]

also, why does your 465B only go down to 0.2V/div? it should be able to do 5mV/div.

[guest]

actually, those pot connections are right in that same area, so it could be a highly resistive "short" to the ACC->RES jack.

[morbank]

Pulled the face plate off, didn’t see any areas where the jack contacts were poking through the tape but I put another layer down anyways and, that was it! I’m no longer getting any pitch fluctuations via the VCF output or when using the VCA output with the external input! I still have no idea what was contacting what, but the issue is gone so it definitely had something to do with the jack contacts.

Whew!

[morbank]

also, why does your 465B only go down to 0.2V/div? it should be able to do 5mV/div.

It does go lower, I just meant that I had set the vertical knob as far down as possible and with the v/div set to anything higher than .2 the trace gets cut off.

[guest]

well, im glad its working. it must have had a very small bit where the RES2 connector poked through, and was causing a bit of current to flow. ive seen this sort of thing before, you cant see any holes, but its still conducting. i like to use a small piece of cardboard or thin plastic (cut up a milk jug) instead of tape as the insulating layer, as its less prone to getting a hole popped in it. im also happy that i can stop wondering what the heck was going on!

[morbank]

well, im glad its working. it must have had a very small bit where the RES2 connector poked through, and was causing a bit of current to flow. ive seen this sort of thing before, you cant see any holes, but its still conducting. i like to use a small piece of cardboard or thin plastic (cut up a milk jug) instead of tape as the insulating layer, as its less prone to getting a hole popped in it. im also happy that i can stop wondering what the heck was going on!

Yes, very glad to have sorted this. Again, thank you for enduring all the back and forth. I very much appreciate the help. And I’m also happy to say that everything is working the way I had hoped in regards to controlling the slide and accent. I’ll try to post a short vid showing it in action before the new year...

[constantG]

I am building this and have a problem. the ERA-V33J561V 560 ohm temp co 3300ppm resistor is now obsolete with nothing even close to it as an alternative. Could someone explain what it does and what should I do as an alternative. tia

[Altitude909]

I am building this and have a problem. the ERA-V33J561V 560 ohm temp co 3300ppm resistor is now obsolete with nothing even close to it as an alternative. Could someone explain what it does and what should I do as an alternative. tia

use 2 1k in parallel and change a resistor value (I forget which one, 27K?)

[guest]

its the tempco to keep the oscillator temperature stable. see what other 3300ppm/C tempcos you can get your hands on, and then adjust the 27k accordingly (as altitude909 suggests).

[constantG]

OK, but looking on mouser, the maximum temp coefficient on all their SMD resistors is 2000 or am I missing something?

[guest]

digikey has some of these still in stock:
ERA-V33J391V

you then reduce R106 to 18k. and props for diying the x0xheart, thats a lot of really small parts to solder!

[constantG]

digikey has some of these still in stock:
ERA-V33J391V

you then reduce R106 to 18k. and props for diying the x0xheart, thats a lot of really small parts to solder!

I'm sorry to say that none of the ERA-V33 are in stock, in any store as they are obsolete. I will keep looking but an alternative solution needs to be found. :(

[guest]

did you check digikey? as best i can tell, they have over 3000 in stock.

[constantG]

I was looking at the UK website and they don't have any, you are correct though, Digikey USA have tonnes of them with $18 delivery :(

[guest]

you can try tacking a 1k through hole resistor from thonk.

[constantG]

Thonk UK have through hole 1k 3300ppm ones, I could use two of these in parallel, what would I need to make R106