Author |
Message |
darkomen
Joined: Jun 21, 2010 Posts: 40 Location: USA
|
Posted: Wed Jun 01, 2011 3:21 pm Post subject:
1v/oct calibrator problem |
|
|
I built the calibrator and it worked great, until i made a little panel and container for it (similar to how ray has it on his site). Once it was finished I went to try it out and it didn't work!? Out of curiosity I hooked my voltmeter up to the panel I created (made from aluminum) and it was reading the correct negative voltage! I'm guessing that that a piece of hardware (probably the out jack) is in contact with the panel, if this is true, how could I remedy it? Or, could it be something else? |
|
Back to top
|
|
|
jordroid
Joined: Jan 17, 2010 Posts: 193 Location: ithaca, new york
|
Posted: Fri Jun 03, 2011 7:59 am Post subject:
|
|
|
Slightly more information would help to diagnose. When you were reading a negative voltage on the panel, where was your red probe and where was your black probe, and what were the voltages you measured?
regards,
jordan |
|
Back to top
|
|
|
darkomen
Joined: Jun 21, 2010 Posts: 40 Location: USA
|
Posted: Fri Jun 03, 2011 8:32 am Post subject:
|
|
|
My red probe was connected to the panel and the black was connected to ground. The voltage measured was the negative of the voltage I expected to come from the jack's output (so if the output voltage was supposed to be 5 volts, the panel measured -5 volts).
Before I put everything together on the panel and in the enclosure I made it was performing as intended. I checked to make sure nothing was accidentally coming in contact with something it shouldn't, but everything looks good, so I'm pretty confused on what is going on. |
|
Back to top
|
|
|
jordroid
Joined: Jan 17, 2010 Posts: 193 Location: ithaca, new york
|
Posted: Fri Jun 03, 2011 9:26 am Post subject:
|
|
|
A jack wired backwards sounds like the likely culprit (if you are using a metal bodied jack which is not insulated from the panel), i've done it a few times, although i don't know why the voltage reading would be negative, unless the probes were reversed. Do you get a beep between the output of U4 and the panel? |
|
Back to top
|
|
|
marvkaye
Joined: Mar 14, 2011 Posts: 225 Location: Fla
|
Posted: Sun Jun 12, 2011 8:14 am Post subject:
|
|
|
I built mine in a small plastic enclosure and had a problem with it initially that I didn't find when I did my voltage and continuity checks prior to installing the ICs.... turns out the switched jack I used for the output was shorted from switch to ring, so without a plug in it it was a dead short across the output. Not realizing it at the time I installed the chips and tried to calibrate. Not only wouldn't it calibrate, I took out my freshly built LM317/337 power supply in the process. Fixed the PS, replaced the jack with an unswitched mono (111) one and tried again. I found that it worked fine when I stepped it manually but I couldn't confirm the auto-step mode with my DMM, as the voltages changed too fast for it to respond, even with the cycle speed turned all the way down. Hooked up to a scope, though, it made a perfect stairstep. After calibrating I found that the variance at each step was never off by more than a fraction of a millivolt, so I'm pretty happy with how it turned out. Lessons learned? Check EVERYTHING for shorts before powering up. And use the scope when checking voltages on a running oscillator, as the DMM is useless in that situation. |
|
Back to top
|
|
|
loydb
Joined: Feb 04, 2010 Posts: 393 Location: Providence, RI
|
|
Back to top
|
|
|
inlifeindeath
Joined: Apr 02, 2010 Posts: 316 Location: Albuquerque, NM
|
Posted: Fri Jun 24, 2011 5:00 am Post subject:
|
|
|
does your multimeter have a continuity test? I use mine ALL the time, super handy for checking if you have shorted traces or good coductivity. _________________ http://www.youtube.com/user/borisandfef |
|
Back to top
|
|
|
marvkaye
Joined: Mar 14, 2011 Posts: 225 Location: Fla
|
Posted: Fri Jun 24, 2011 6:25 am Post subject:
|
|
|
Quote: | How do you check for shorts without powering up? |
As inlifeindeath mentioned, setting your DMM to continuity check (where it beeps when the leads are shorted together) or at least to the lowest resistance setting, with the item to be tested not under power and with the chips removed, it's easy enough to check for shorts on the IC sockets between adjacent pins or on jacks and such by taking readings between their terminals. Doing this with a schematic available will at least give you some answers about resistance readings that show up, most times they're pretty easy to track. You can also check that switches are wired properly, etc, etc. I was pretty embarrassed by not finding my jack short during my intial checks... it should have been part of the process. I guess I just missed it. After the shorts check is done you can power up the unit without chips installed and verify that all the voltages are proper. If you go to the MFOS website and look under his Synth DIY 101 link, down at the bottom you'll see Troubleshooting 101... it should be called "PRE-troubleshooting..." as if you follow those directions it's likely you'll have a working unit that doesn't actually require troubleshooting when it's powered up, as most of the problems will have been located before you get to that point.
You can see what my calibrator looks like in the following message:
http://electro-music.com/forum/post-333290.html#333290
While we're on this topic I thought I'd mention that I can calibrate my unit to provide extremely accurate 1V/step outputs but noticed that it does start to drift a bit and needs to be tweaked every now and again. I spoke with Ray about this and he confirmed that the drift is not unusual and is likely related to a minor fluctuation in the zener voltage as it heats up. He's apparently thinking about changing how the 9V is developed by replacing the zener with a more robust regulated circuit, but I imagine it's a low priority and just another of many things on his to-do list for the future. In the meantime, knowing that the drift is there it's easy enough to tweak the trimmers for accuracy just before using it to calibrate your VCO to insure the best results.
I'm thinking about removing the zener and R2 and injecting a regulated 9V from a benchtop PS into the place on the PCB where 9VZ starts, at the cathode pad for Z1. This should keep the 9VZ voltage constant throughout the rest of the circuit and eliminate the drift... at least in theory...
_____________
<Marv> |
|
Back to top
|
|
|
loydb
Joined: Feb 04, 2010 Posts: 393 Location: Providence, RI
|
|
Back to top
|
|
|
kvitekp
Joined: Feb 24, 2012 Posts: 22 Location: Santa Clara, CA
|
Posted: Fri Feb 24, 2012 7:23 am Post subject:
|
|
|
To eliminate the drift I replaced 9.1V zener with LM4040CIZ-10.0-ND part which is a Precision Shunt Voltage Reference with .1% tolerance and pretty low temperature coefficient (100ppm/C) in TO-92 package.
The output voltage is rock solid now. Before the mod I had to calibrate calibrator every time I used it since where I live temperature change of 20 degrees day to evening is a norm.
BTW, similar mod works great for transistor Vbe matching precision power supply.
/Peter |
|
Back to top
|
|
|
|