Way OT!!!! How to kill some mV leakage? Trust me it's OT!

[BillyJ]

Well.....I'm not sure.

I still don't know if that 1M resistor is masking any mV leakage or just providing me with the correct reading.

If it is correct than I either do not have a problem or the mV leakage is not the nature of the problem.

If that is true I would guess either there is a mA leakage or this guy never saw any leakage on his because he never tested it.
He could very well have only tested mine with a meter saw a mV reading and belived it would not work never having hooked it up to the watch.
From reading some email with him I am getting this impression. Which is a pisser because I told him from the get go that I did not belive my meter was correct.....
I iwll be pretty upset with this guy if I find this to be the case. I have damn near pleaded with him to try one with the watch.
His working circuit is a breadboard with glue on it. I can see why he wouldn't want to but damnit if these things work and he has had me embarss myslef in front of the whole fx community for nothing I really will be displeased with the whoile shebang.

Can you tell me Peter or Paul (or anyone) if that 1M resistor across the meter leads will give me an accurate reading?

I am almost ready to deal with this properly and bust open his damn precious timer.
I just want to be pretty sure of what the real issue is.
Sorry if all we are debugging here is my lack of sense....

Once again, you gentlemen are awesome! THANK YOU.

[Peter Snowberg]

No worries Billy.

You keep mentioning leakage, but I think you're headed in the wrong direction. The stuff you want to test cannot be tested with the meter directly because the stopwatch is running off of almost zero current.

This is what I would do:

* Test the wires going from the opto to the stopwatch. Short them and verify they're working like the pushputton.

* Connect the wires to pins 4 and 5 of the opto. Connect a 1K resistor from the + side of a 9 volt battery to pin 1. Connect pin 2 to the battery - and then remove. See if that acts like pressing the button.

if not....

* Swap the wires on pins 4 and 5 and try the battery step above again.

If you don't get action on the stopwatch, your opto is the problem. Forget measuring anything here. Watch how it operates in circuit since this is a basic situation you can monitor. The stopwatch starting and stopping is the best meter you can get.

A meter with a diode test is useful for testing the opto. Connect the red meter lead to pin 5 and the black to pin 4. Turn the meter to diode test and connect the battery through a 1K resistor. You should see nothing without the battery and with the battery connected the voltage drop should read around 0.5 volt. Make sure the opto works.

I'm guessing the stopwatch is an LCD unit that runs from 3V. Close? If so, one of the switch wires will be a ground or a power wire and the other will be a CMOS input with a pull resistor that could easily be 22 MEGS. Anything you attach to that wire will influence it. Make that wire long and local radio stations will influence it.

You might be able to use a JFET instead of the opto. RadioShack sells MPF102s which I think will work here. This is just an idea. You milage may vary. Try connecting the stopwatch wires to the drain and source of the MPF102.  Connect the 9V minus lead to the MPF102 source. Connect a 10K resistor from pin 3 of the 555 to the MPF102 gate and connect a 1 meg resistor from the gate to the source. Try it out. If that doesn't work, swap the stopwatch connections.

Good luck,

-Peter

[Paul Perry (Frostwave)]

I have thought some more..
The opto has a transistor output. So the collector of the transistor needs to go to a load resistor (say 4k7) to the +9 power (or whatever your 555 is running from).
Now when the opto is OFF, the output (from the collector, not the base, as drawn in the circuit!)
will be not far from +9 (or whatever the supply is), and when the opto is ON, the output falls to
a couple of diode drops above ground..
Now this isn't good enough for the watch, that is wanting to go from +1.5 to much less, while you have it going from +9 to +0.8 or so. BUT!!if you put a two resistor divider from the collector to ground, you can divide down both the 9 (when off) and the 0.8 when on. For example, with a 110k from the collector to a 22k and then the 22k to ground, the junction of the resistors outputs approximately 1.5v when OFF and only 0.16 when ON, which surely must be low enough to turn off the watch.
Some experimenting is called for.. be sure not to ever send more than 1.5 to the watch, in case this is fatal for it.

[Rob Strand]

I'm not sure I agree with Paul.  It's not a good idea to feed +9V into the stop watch if the stop watch is running from 3V - basically you will fry it.

From my understanding the stopwatch just wants a closed contact to start - the contact should be like a switch with no potentials.  It is unclear whether the contact must be released a sort time after, or if the watch ignores a held on switch - this is probably an important consideration to guarantee it working .  It is highly like the potential across the switch is polarized this means you have to get the switch connections around the right way when using a transistor switch (as it is in the opto).

From the schematic it appears you are connecting the output across the base and emitter - this won't work in most circuits.  Usually you want it across the emitter and collector..

It looks like to 470 ohm resistor could be a little low in value, it probably won't affect the operation but it will waste power on the timer.

The way the 555 is triggered looks a bit dodgy, however, you might be able to use this method depending on the application - it could even be a desired behaviour in the application.  Basically the 555 monostable resets the timing after each trigger so in the current circuit the output will go hgh starting from the *first trigger*, and will go low a the programmed  amount of time after the *last one* - if you keep getting triggers it will not time-out.

[Paul Perry (Frostwave)]

Rob, I didn't say feed 9v to the watch! i distinctly said, DONT do that.. what I say, is that the easiest way to get wht seems to be wanted by the watch ( a voltage that is either 9v or "very low"), is to get the opto set up to output 9v (off) or a few hundred mv (on) and then divide down this output voltage by 2 resistors in a chain to ground.

Now, I know this isnt how I would WANT to do this.. but, given the stuff we have, and given that I don't even hav the specs for the watch !!, this is the best I can think of, to get the thing going without specifying more parts.

BTW, I cn't see he ckt nymore either.. but I saved it

[Rob Strand]

Rob, I didn't say feed 9v to the watch!

You will have to forgive me Paul, this morning was the first time I've been able to see the schematic and I blasted through the thread a little too quickly and missed some of the finer points.

I suspect the off and on voltages Billy mentioned are a bit of a furfy and that the watch has it's own pull-up on the input.  If the watch really needs the input to be pulled down V-close to 0V then a BJT output might not be the way to go, perhaps a MOSFET would be better.  If the unit is self powered, and not grounded, or the watch isn't grounded then there's no real need for isolation - it's already isolated.

(IMHO, it's not a good idea to develop something without a full data sheet and/or a real unit, there's too much guessing - and that leads to trouble!)

[Paul Perry (Frostwave)]

hey, no offence taken, Rob! I don't doubt you are more 'on the case' than I am And I'd like to know what these watches do, in case I ever have to time my mate's greyhound! (a poor sixth at Wangaratta tonight).
Things could be worse, at least I havn't suggested putting a LDR across the watch..

[BillyJ]

Whoa I am in so over my head with this. I cannot thank you guy enough you are awesome. I am going to call this guys bluff and getthe damn thing here. I know either he never even tried the damn things and they will work fine or the problem lies in a incorrect component.
It has to something very simple because I am following it to a T.
I know about the pinout issues but this is what he tells me is working for him.
i will get that damn box from him figure out what the problem is and give him a rash over it. I will get back to you all and let you know.
Man oh man from now on if it doesn't make a noise I am not going near it.
Now I need to get into my chapter about Ohm's law....yikes!
Thank you all so very much for trying to help me. I wish like heck I could follow along.  :cry:

[Rob Strand]

And I'd like to know what these watches do, in case I ever have to time my mate's greyhound! (a poor sixth at Wangaratta tonight).

Funny you should mention that, it just occured to me I designed multichannel/multi-stopwatch timing system for an old colleage of mine.  It was for model boat racing. You had a start botton (synched to the starting gun) and individual buttons for each boat.  I used CMOS 4016 switches for the stopwatch switches.  I remember these worked because my friend had tested them out before he asked me about it.

[Rob Strand]

Whoa I am in so over my head with this.

Billy, don't give up yet.  If you have a real unit, and a whole set-up, you can play around with it and get a feel for the problem.  You will probably need to do a few experiments and before you know it you will be back on your feet. It makes life soooo much easier - and it eliminates that stuck in limbo feeling you probably have at the moment.

[Peter Snowberg]

Don't worry about it all making sense right now Billy. It all kinda gets absorbed and eventually the dots connect themselves and the light goes on. I hope you don't give up on non-musical electronics. There is good learning to do everywhere and electronics is VERY diverse. My background is mainly digital and that didn't help as much as I thought it would when I started digging into analog. When it comes to RF electronics, I might as well be a plumber.  :?

Concentrate on boxes if you wish, but please don't neglect the other stuff. The idea is to infuse knowledge from those other areas into the effects world to make new things.

You have lots of people here that will gladly work through every question you have. Don't worry about your current level, just say "Hey!, make it simpler" and we will. Everybody doing more complex stuff has gone through exactly the stage you're in, but many of us didn't have internet resources like this. That would have been sweet.

One tip: Remember that methodology is almost as important as the knowledge itself. You can do the 'wrong' thing in a creative way and invent something unexpected.

Check out the middle of this page where is says "standard solid state relays". This is what I would try.

http://www.mouser.com/catalog/615/126.pdf

I would go for one of the units with the output type listed as "1a" (normally open). They use MOSFETs instead of biploar transistors (NPN/PNP). The stopwatch is made using MOSFETs because they use almost zero power when switched off, which is perfect for battery devices. If you use the same animals, everything gets much easier to work with (of even possible for that matter).

It doesn't matter what your question is. If you see anything that doesn't make sense, it's pretty easy to adjust the answers until they do. Just ask.

-Peter

[Paul Perry (Frostwave)]

that's terrific advice from Peter Snowberg, it should be stuck on refrigerators everywhere, along with the AA prayer! Whenever I am having trouble, I take comfort that I only learn things when I try to do stuff without knowing everything I needed to.
"the man who makes no mistakes makes nothing".

[BillyJ]

You guys are the best! Thanks so much for the encouragement. I won't skip out on the other stuff Pete. I am gonna stick to it.
I am not going to try to learn it all on this kooky timer that is for sure.
If I can't change it right now I can accept that.  :wink:
I will get back to this when I can see just what in the world really separates the two....Thanks again you guys very kind of you.


Oh wait...Paul, Pete answer me this one thing please...
If I have a 1m rewsistor across my meters leads is it masking any milli voltage that could be there?

[Paul Perry (Frostwave)]

Normally, having 1meg across a meters leads when measuring voltage in a circuit won't make any difference, unless you are working with very high impedances. In any case, you could leave a 1 meg resistor across the input to the watch.. without having details of the watch, I can't say anythng much about what will affect it.

[Peter Snowberg]

 Thanks Paul!

Billy, always try to remember the quote Paul just gave:

"the man who makes no mistakes makes nothing"

That is so amazingly true.

Einstein had I quote I always liked, "If we knew what we were doing, why do think we would call it re-search?"  :wink:

-Peter

[BillyJ]

Got the timer! The dude had his opto flipped upside down.
I had complained originally that the pinout he wrote down was wrong.
When talking to him on the phone he said "Oh yeah your right I did goof that, here is what the pinout really is." At which point I think he told me the pinout as if the opto was upsidedown on his breadboard (which it is).
SO I have learned a ton. If not exactly about electrons then about how to get all the things you need for a project up front with no handdrawn schematic, without a way to testthe function....oh I could go on and on.
Sooooo... I am going to be drilling some holes, Cleaning with some acetone, hoping I can get a strong connection using the optos pad to hold the opto pin and a wire...

Peter I read you talking about a chemtronics conductor pen. I know it is a dumb question and pretty sure the answer but that stuff can't be soldered to can it?

I was also thinking I could desolder the one opto pin with the wire attached, cut out some copper from between the opto pad and tghe wire pad and then just bend the wire over to the next pin....
It would save having to drill another hole which might keep it looking more or less clean. Like there were no errors. I am just not at all interested in redoing the layout and making new boards.


Well in any event I will be fixing these damn things and as soon as I get paid the paltry amount I agreed to do it for I iwll be paypalling Aron a little chedder to help with the board or whatever he pleases.
Thank you all very much for the help. I am sorry I am so much of a lunkhead I could not see the problem for myself even when it is pointed out to me.

I cannot thank you all enough. I might have had a very doifferent experience with this mess without this forum. Sorry it was so offtopic.
Anyone want to buy a really cool Slalom Skateboard Timer? LOL!!!

[Peter Snowberg]

Peter I read you talking about a chemtronics conductor pen. I know it is a dumb question and pretty sure the answer but that stuff can't be soldered to can it?

That's not a dumb question at all.

I wish you could solder to conductive ink, but I've never heard of one that will let you do it. You can however terminate your soldered connection and then draw a trace to it. If there is nothing to solder a wire to in order to anchor it, you can even anchor it with epoxy and then draw the trace over that.

With the change you need to do, I would try to just cut the trace(s) right in front of the opto pads that need changing and use small jumpers on the bottom. Resistor leads and needle nose pliars come in handy here. That avoids using the drill again and as you mention, the resilt will look cleaner.

Some of the hardest problems are the ones that stare us in the face. The more you fool around with anything the easier it gets to miss "small" errors. I've spents hours and hours scratching my head only to finally mutter a big "Duh!!!!" once I saw was I was overlooking.

I've always loved the trueism: "Hide in plain sight"


take care,
-Peter