Author |
Message |
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sun Jun 29, 2014 11:56 am Post subject:
|
|
|
So pin 7 U3 I am getting -12v/+5v scope reading. Same reading for D18 anode, cathode is +5v, pin 14 U1 is also +5v. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sun Jun 29, 2014 1:05 pm Post subject:
|
|
|
Images to support previous post.
Description: |
|
Filesize: |
2.4 MB |
Viewed: |
345 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
1.28 MB |
Viewed: |
425 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Sun Jun 29, 2014 7:42 pm Post subject:
|
|
|
I would be interested in the pulse WIDTH of the signal at U3B pin 7.
Per Ray's description, R29 is supposed to lead to some hysteresis to the comparator, I would expect this to extend the pulse width, but the scope shot is mighty narrow at the time scale you're using.
You might want to verify that R29 is indeed 1M (apologies if you already have and mentioned it previously, I was hiking around botanical gardens all day ). You might also check that it has good continuity (no cold solders, no shorts) to the two pins of U3 (5 & 7) and the two sides of C2 that it connects to.
What does the clock pin 14 on the 4017 look like when you are able to successfully trigger it with the step switch? What peak voltage, what pulse width? |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Tue Jul 01, 2014 12:52 pm Post subject:
|
|
|
A distraction. I looked into poor vertical trace on oscilloscope, in short I think my scope has a fault which is why the pulse(s) in the scope shots are so faint. I believe its still just usable. I ran it in test mode and got hardly any vertical trace. The height of the vertical trace is important not necessarily its strength?
I will confirm pulse widths and peak voltages results in next post. I did check the cathode of D18 with the scope and got +5v but I want to double check it.
Recap. On cold solder I have swapped out: R29,C2,R23,R28,R32.D17,D18. This leaves D16, I have also swapped out the socket of U3. Plus I checked all replacement parts with multimeter before soldering.
I'm not giving up hope javascript:emoticon('') but seem to be running out of options in U3b area. Am I right in thinking the issue is why I am only getting -12v/+5v on pin 7 U3 and therefore missing or have lost +7v on this output? |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Tue Jul 01, 2014 1:59 pm Post subject:
|
|
|
Cato wrote: | A distraction. I looked into poor vertical trace on oscilloscope, in short I think my scope has a fault which is why the pulse(s) in the scope shots are so faint. I believe its still just usable. I ran it in test mode and got hardly any vertical trace. The height of the vertical trace is important not necessarily its strength?
|
Yes, height gives the peak voltage, and width gives us a time base to know if the pulse is long enough to work. I don't know the exact requirements of the 4017, but they're in the data sheet.
Cato wrote: |
I will confirm pulse widths and peak voltages results in next post. I did check the cathode of D18 with the scope and got +5v but I want to double check it.
Recap. On cold solder I have swapped out: R29,C2,R23,R28,R32.D17,D18. This leaves D16, I have also swapped out the socket of U3. Plus I checked all replacement parts with multimeter before soldering.
I'm not giving up hope javascript:emoticon('') but seem to be running out of options in U3b area. Am I right in thinking the issue is why I am only getting -12v/+5v on pin 7 U3 and therefore missing or have lost +7v on this output? |
Yah, whatever is causing it to not be able to go above 5V is my current operating theory. That's enough to reset the FET (so we have oscillation) but perhaps not enough to clock the 4017 with a 12V supply.
min pulse width, btw, for a 10V supply is 90ns, and the higher the voltage the shorter the pulse can be. For a 15V supply, it's 60ns, so I'd guess 12 is probably around 80ns.
It looks like the chip wants a "high" signal to be at least .7 of the supply voltage, so we'd need to exceed 8V to get the pulse to go. That's partly why I'm curious what the pulse looks like when you successfully use the step button--what is the peak you get then versus when it's not clocking from the oscillator. |
|
Back to top
|
|
|
prgdeltablues
Joined: Sep 25, 2006 Posts: 222 Location: UK
Audio files: 12
|
Posted: Wed Jul 02, 2014 3:56 am Post subject:
|
|
|
That would be my thinking too.
A thought that occurs to me is, if the U3 circuit resets too quickly, the pulse may not only be narrow, but also may not have time to reach +12V. From Ray's notes: "The reset pulse from the comparator is quite narrow due to the fact that Q11 turns on so quickly and brings U3-A's output below U3-B's threshold. But that squirt of charge into C5 keeps Q11 on a bit longer".
C5 is shown as 1nF. I wonder if that is in fact marginal. I'd suggest swapping it for a 10n cap (actually, just put a 10n in parallel, no need to desolder). It's possible that might cause problems at high clock speeds, but if it gets the pulse clocking the 4017 we'd be getting somewhere! (I had a similar problem recently with a reset pulse to a 4017, where the circuir called for a 10pF cap, but I found I needed 100pF to get it to work, for this reason.)
Peter |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Wed Jul 02, 2014 1:08 pm Post subject:
|
|
|
Thank you both for input need to get back to you asap.
I am still trying to find the time to investigate your recent feedback. I scoped D18k as it goes straight to pin 14 of the 4017. In start mode this gave +5v, with the "forced" sequencing via pin 5 U3 I get +7v. Will look into swapping out C5 1nf for a 10nf. |
|
Back to top
|
|
|
Chartreuse-J
Joined: Jan 21, 2014 Posts: 8 Location: Dixie
|
Posted: Thu Jul 03, 2014 2:00 pm Post subject:
|
|
|
I had this exact same problem with my sequencer...
Thread here...
http://electro-music.com/forum/topic-60862.html
I finally gave up...
I think Ray needs to look at the parts spec or something...
Oh and no its not a ground issue. |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Jul 03, 2014 2:23 pm Post subject:
|
|
|
Chartreuse-J wrote: |
Oh and no its not a ground issue. |
But I was SO CERTAIN
Sorry to hear it, it always sucks to have a problem beat you. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sat Jul 05, 2014 12:07 am Post subject:
|
|
|
Thanks for input Chartreuse-J, I have looked at your post several times over the past couple of months.
I tried Peters idea with C5 :
C5 is shown as 1nF. I wonder if that is in fact marginal. I'd suggest swapping it for a 10n cap (actually, just put a 10n in parallel, no need to desolder). It's possible that might cause problems at high clock speeds, but if it gets the pulse clocking the 4017 we'd be getting somewhere! (I had a similar problem recently with a reset pulse to a 4017, where the circuir called for a 10pF cap, but I found I needed 100pF to get it to work, for this reason.)
No change.
I am wondering if its a spec variation in IC brands, that has tolerence(s) on the edge of the limits, which is why it works for some and not others? I found this curious post
http://electro-music.com/forum/topic-49292.html
I have a TI 40106 in use.) javascript:emoticon('')javascript:emoticon('') |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sat Jul 05, 2014 12:33 am Post subject:
|
|
|
I am going look at specs for TL072, check what I have used against other brand specs. Can't see why my issue would be the 40106 at this stage, though comparing a 'TI' to a 'Fairchild' I see a lot of differences in the specs.
Clutching at straws!!javascript:emoticon('') |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sat Jul 05, 2014 4:23 am Post subject:
|
|
|
SUCCESS IT WORKS! .
Looked at u3b again,soldered a 22 swg wire to pin 5 , licked my fingers and pressed on exposed end of wire - got automatic sequencing. So I am supplying enough charge to get it working.
What is pin 5 connected to? R29,R23 and C2. Started to think do these component values need tweaking? Tried R29 first increasing and decreasing capacitance, a bit of erratic sequencing but no sound result.
Then looked at C2 again (see page 2 of thread) thought I'd touch a higher value cap 1000pf (code 103) against the existing 10pf and IT WORKED!
Swapped out 10pf for 1000pf and as far as I can visually tell everything works. I'm now going to scope test the outputs and breadboard a basic VCO to see if I can voice the notes.
Considering building the Thomas Henry VCO as I had quick success with his VCA design using veroboard.
Thank you for the guidance,advice and interest given by all of you. Especially to Elmegil - you have taught me a lot. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sat Jul 05, 2014 7:15 am Post subject:
|
|
|
Not the greatest scope images... first shot is D18k and pin14 4017. It shows about +10v. Second shot shows D18 anode, about +10v and -12v. The volt/div is 0.5 and the time/div is 10ms.
One last question? Shouldn't I be getting scope readings from the gate and trigger outputs? (as shown in schemo) I am getting readings from all 3 CV outputs only, +4 to +6v.
Description: |
|
Filesize: |
2.88 MB |
Viewed: |
330 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
2.89 MB |
Viewed: |
295 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Sat Jul 05, 2014 9:42 am Post subject:
|
|
|
Cato wrote: | Thank you for the guidance,advice and interest given by all of you. Especially to Elmegil - you have taught me a lot. |
You're quite welcome. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sat Jul 05, 2014 10:11 am Post subject:
|
|
|
It's still got issues.('') - I know CV outs are working as I got scope peaks and meter voltages. Built a simple 555 VCO, to voice the pulses.
Have had 30 minutes of fun actually listening to it but only with the CV outputs, portamento works, step selector switch works as do all start/stop, step and reset buttons, CV out range selector works and so does the clock.
Two remaining issues, no signal/pulse on gate and trigger outputs - will buy non TI 40106 ic in reference to http://electro-music.com/forum/topic-49292.html
Gate switches do not turn off individual gates, back to the schematics.
Still wondering why C2 cap on U3b was the cause of the non sequencing. |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Sat Jul 05, 2014 11:51 am Post subject:
|
|
|
Cato wrote: |
Still wondering why C2 cap on U3b was the cause of the non sequencing. |
Should look into descriptions of how an op amp wired as a comparator works... (as opposed to a dedicated comparator). I'd have to review myself to try and figure it out. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sun Jul 06, 2014 11:32 am Post subject:
|
|
|
No more issues it all works
The last fault, no gate and trigger out, was caused by a the GCOM wire being loose on the first gate switch. Now re-soldered and everything works correctly.
Once again thanks for the help and if anyone's interested you can see it in action here http://youtu.be/a6GI_RQREEE |
|
Back to top
|
|
|
Chartreuse-J
Joined: Jan 21, 2014 Posts: 8 Location: Dixie
|
Posted: Mon Jul 07, 2014 10:50 am Post subject:
|
|
|
Kudos, glad you got it working...
I'd be curious to know what voltages you are getting respectively from the gate and trigger outs and CV.
On a side note...
So all it was just a simple swap of the C2 1pF to a 1000pF? Interesting, I'll bet Ray should be informed so he can change his specs.
But why would his work with a 1pF and yours does not, was it a typo?
In addition..
I am pulling all the ICs out of mine and reordering new ones, would you care to share the exact part number / MFr Part # for all the ICs.
For example.....there are like 7 different TL072 in there from Texas Instruments all with different Input offset voltages listed on Mouser. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Tue Jul 08, 2014 12:05 pm Post subject:
|
|
|
The Gate and Trigger both show 10v using a scope.
Typo error in previous post of mine. A 103 cap is 10 add on the 3 zeros and you get 10000pf not 1000pf as I stated earlier.
This is now irrelevant as I got to thinking 10000pf is a big value change. Took out C2 again and tried a 22pf no luck, I then tried a 22pf in parallel with a 10pf. Success it auto sequenced, so 32pf is fine. Which brings it a lot closer to Rays' spec of 10pf. I could only guess at the reason for this, slight tolerance variation in ICs? C2 at 10pf near its limit? I wonder if any one else out there had same issue?
The chips I have used are ;
ST electronics TL074CN GK735299
TI CD4013BE 13AMS42 E4
TI CD4017BE 1AD59LK E4
TI TL072CP 18MDSHY
TI CD40106BE 12AMS30 E4
Have you got a scope? Scope testing the outputs of the ICs will tell you if they are working, don't know your level of expertise. I'm definitely no expert! I would try leaving the Ics in and try C2 first. |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Sun Oct 05, 2014 5:46 am Post subject:
C2 again Subject description: changed cap value |
|
|
Four months on and the start/stop has failed. I had a 103 ,10000pf, cap handy so soldered this to C2. WORKS. |
|
Back to top
|
|
|
Chartreuse-J
Joined: Jan 21, 2014 Posts: 8 Location: Dixie
|
Posted: Wed Oct 15, 2014 4:17 pm Post subject:
|
|
|
Ok,
Heres an update....
Put all the new ics in the unit and tried powering it on, still no dice.
Removed the cap from location C2 and replaced with a 50pf ceramic cap in there and it now works flawlessly! Gates triggers portamento, clock speed all work! Everything does!
Yes it is alive finally!
I even sequenced an old oscillator with it and it sounds great!
On a side note I think Ray should update his schematics on this sequencer, or at least chime in as to why this mod works.
Also I have gotten a maximum of about 8v out of the CV outs and 4v when in low mode. Triggers are 10v and gates are 10v, perfect for eurorack levels! |
|
Back to top
|
|
|
Cato
Joined: May 29, 2014 Posts: 44 Location: uk
|
Posted: Fri Jun 03, 2016 12:45 pm Post subject:
Still working |
|
|
still works fine with no changes since 2014 |
|
Back to top
|
|
|
|