Is the magnet soldered to a PCB or are there wires around anywhere?
A picture might help.
B
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
macOS Rotary Switch to Variable Timing Circuit?
- Thread starter Sdashiki
- Start date
- Sort by reaction score
Is the magnet soldered to a PCB or are there wires around anywhere?
A picture might help.
B
It has 2 connections on a circuit board on the opposite side. Solder joints I can probe.
the magnet is right below the far right arrow, green coiled
on the opposite side of this photo is a circuit board like whats on the left. the magnet has 2 joints just like that where you can follow their connection to the other components if you need to.
the magnet is right below the far right arrow, green coiledon the opposite side of this photo is a circuit board like whats on the left. the magnet has 2 joints just like that where you can follow their connection to the other components if you need to.
To measure current you need to effectively break the circuit and then bridge your break with the multi meter. In a situation like this, it usually means either de-soldering or cutting a wire.
Why not have the timing circuit activate a solid state relay then, isolating it completely from the electromagnet circuit?
Or alternatively trace components and see if you have something like a "current limiting resistor" directly in line with the magnet/motor and just measure voltage across that.
B
Sigh, no wonder.
I had the DMM at AC and not DC...jeez.
Touching the - and + end of the circuit, running the 3v power thru it gives me 26mA
I had the DMM at AC and not DC...jeez.
Touching the - and + end of the circuit, running the 3v power thru it gives me 26mA
What about a 555 chip circuit built to do a 1second pulse.
Then, based on the math formula for the res/cap combo...replace the res with a pot of the same strength.
Then, as I turn the pot, the 1 second pulse becomes 1/2, 1/4 and so on...am I correct in this simple assumption?
if so...what is the precise wording for what Im looking to research, I find PWM and one-shot monostable but cant decide since looking the diagrams is mostly doodles to me when it gets complicated.
Then, based on the math formula for the res/cap combo...replace the res with a pot of the same strength.
Then, as I turn the pot, the 1 second pulse becomes 1/2, 1/4 and so on...am I correct in this simple assumption?
if so...what is the precise wording for what Im looking to research, I find PWM and one-shot monostable but cant decide since looking the diagrams is mostly doodles to me when it gets complicated.
What about a 555 chip circuit built to do a 1second pulse.
Then, based on the math formula for the res/cap combo...replace the res with a pot of the same strength.
Then, as I turn the pot, the 1 second pulse becomes 1/2, 1/4 and so on...am I correct in this simple assumption?
if so...what is the precise wording for what Im looking to research, I find PWM and one-shot monostable but cant decide since looking the diagrams is mostly doodles to me when it gets complicated.
The Wikipedia page on the 555 isn't bad. http://en.wikipedia.org/wiki/555_timer_IC I've got the Forrest Mims Engineer's Mini-Notebook mentioned later in the article somewhere around here.
I think the problem with this approach is that getting the smaller steps (e.g. 1/30 to 1/60 ...) would be very hard to get with a linear pot. Maybe a logarithmic pot? If you ended up with an array of passives you're back at having size issues.
The timing will also not be quite as precise/reliable as the MCU approach.
B
If I am looking to drop the 555 timing from 1 second by half/double at equal intervals around the 300° of the pot (1/2, 1/4, 1/8 half over and over)...is that a log pot?
If the 555 changes timing based on the resistance vs capacitance numbers
time = R x C
1sec = 1Mohm x 1uf
Replace the 1Mohm with a pot...and when dialed to 500k, it works out to 1/2s, so on and so forth.
If the 555 changes timing based on the resistance vs capacitance numbers
time = R x C
1sec = 1Mohm x 1uf
Replace the 1Mohm with a pot...and when dialed to 500k, it works out to 1/2s, so on and so forth.
Yes. Otherwise think about it this way for a linear pot:
1s = 300 deg
1/2s = 150 deg
1/4s = 75 deg
1/8s = 37.5 deg
1/16s = 18.8 deg
1/32s = 9.4 deg
1/64s = 4.7 deg
1/128s = 2.3 deg
etc.
IIRC log pots are not quite logarithmic or very accurate and more used for things like setting volume. (EDIT: I'm also not sure that you could find one that actually covers the dynamic range you want. Even 8 bits might be tricky. Some of the ones I found quickly on Digikey only had about a 40x range ~6 bits from min to max range of R.)
B
Last edited:
Yea, the curves suck on the log ones ive tried.
Though whats this about putting a resistor with a linear to make it smooth out to more of a log curve better than the crappy log ones now? They seem to make 1M pots with +/- 10% for a reasonable price.
Maybe I should think about using the 555, a rotary switch, and a resistor bank all tied to a single 1uf cap.
That'd be...what...1 small cap and then 1 resistor per setting on the switch. Seems simple enough now to me, though making a small enough board to fit in the space...may prove a challenge.
Though whats this about putting a resistor with a linear to make it smooth out to more of a log curve better than the crappy log ones now? They seem to make 1M pots with +/- 10% for a reasonable price.
Maybe I should think about using the 555, a rotary switch, and a resistor bank all tied to a single 1uf cap.
That'd be...what...1 small cap and then 1 resistor per setting on the switch. Seems simple enough now to me, though making a small enough board to fit in the space...may prove a challenge.
Last edited:
As I suggested earlier in the thread find an MSOP 555 and use a pre-made MSOP test board or even an MSOP-DIP converter as your board and you can keep the size way down.
You wanted something like 8-10+ bits of range. Are you really going to get that with this approach?
B
If its a single cap and a resistor bank, id get all the range the ratio of the C to R values give me. Right?
And the 555 chip plus about 10-12 resistors 1/4w, takes up little space.
And the 555 chip plus about 10-12 resistors 1/4w, takes up little space.
Last edited:
Correct, and then you're back to your original SP12T switch for the resistor bank. You might want to plan on two components per element to give you a bit more wiggle room for realizing your R values to get the timings right. Substitute some zero ohm resistors or jumpers if you don't need a given component.
B
Well a resistor bank to a 555 for timing is far better than what I was doing with a bank before. See the circuit right now uses res/cap for timings, and I was attempting a res bank to feed the cap for different times. But, every circuit is slightly different, meaning the res/cap values change making reproducibility, nil.
This way, I can just gut it, replace with this setup and go.
This way, I can just gut it, replace with this setup and go.
I must be missing something. You are still depending on an RC time constant for all your timing with this approach.
All the 555 does it turn that into a nice square pulse. You're still at the mercy of the values and tolerances of the discrete components and their temperature coefficients for the actual timing.
The nice thing about the MCU approach is that you could feed it with a nice quartz clock and be sure the timing was dead nuts on.
B
Boarduino or Ardweeny
Check out the Boarduino , an Arduino clone that is much smaller (7.5cm x 2.0cm, and can be configured to be as "short" as 1 cm tall) than the current Arduino Uno board. I've used this, and like it. That should fix your space problem.
And if I'm fully understanding your potentiometer problem, this should fix that, too. You can use code to set the resolution and range of your pot pretty easily. (HINT: use map() on the pot's analog pin's reading.) Then, use the pot reading to set the pulse and delay in milliseconds.
The one limitation of Boarduino vs Uno is that it cannot supply power beyond its pins (5V DC, 600mA max I believe). The Uno has a separate power bus (is that the right word?) that can supply power to other things beyond the pins. But, I don't see that being a big problem for your application.
I'm not sure of budget restraints, but the Boarduino is about $18. You will also need a chip to program it if you get the DC version. I use FTDI Friend, which is $15 and can be reused over and over. The USB version has the programmer on it already, but you then have to power the board with USB. Tradeoffs...
I'll admit, I've got a seat on the Arduino bandwagon right now, so every project that comes to mind I think how to get the Arduino to do it. It may not be the best way, but it sure is quick and simple.
Disclaimer: I'm no EE, just a hobby tinkerer.
PS: I somehow overlooked the Ardweeny! It's even smaller--the footprint is the IC. I haven't used it, but from what I understand it works just the same as Boarduino except there is no power adapter or voltage regulation built in. Maybe I need to get one and find out!!!
Check out the Boarduino , an Arduino clone that is much smaller (7.5cm x 2.0cm, and can be configured to be as "short" as 1 cm tall) than the current Arduino Uno board. I've used this, and like it. That should fix your space problem.
And if I'm fully understanding your potentiometer problem, this should fix that, too. You can use code to set the resolution and range of your pot pretty easily. (HINT: use map() on the pot's analog pin's reading.) Then, use the pot reading to set the pulse and delay in milliseconds.
The one limitation of Boarduino vs Uno is that it cannot supply power beyond its pins (5V DC, 600mA max I believe). The Uno has a separate power bus (is that the right word?) that can supply power to other things beyond the pins. But, I don't see that being a big problem for your application.
I'm not sure of budget restraints, but the Boarduino is about $18. You will also need a chip to program it if you get the DC version. I use FTDI Friend, which is $15 and can be reused over and over. The USB version has the programmer on it already, but you then have to power the board with USB. Tradeoffs...
I'll admit, I've got a seat on the Arduino bandwagon right now, so every project that comes to mind I think how to get the Arduino to do it. It may not be the best way, but it sure is quick and simple.
Disclaimer: I'm no EE, just a hobby tinkerer.
PS: I somehow overlooked the Ardweeny! It's even smaller--the footprint is the IC. I haven't used it, but from what I understand it works just the same as Boarduino except there is no power adapter or voltage regulation built in. Maybe I need to get one and find out!!!
Well, if nothing else, at least 555 chips are cheap. If the project works, great. If it doesn't, the OP isn't out much.
Exactly. And why should I rely on resistor timing if thats what I am replacing?
Because there are at least 16 types of circuits that all do the same thing, but using different parts as the years went on. The mechanics never changed, so by removing all the original res/cap and variable res components with something "standardized", I can do it on any of the 16+ types.
Check out the photo I posted earlier, the circuits on the left are about the size of the space.
This analog replacement approach seems more doable, cheaper and less risky to my brain. However, the programmable board in the future is a definite possibility depending on how well this holds up to real world use.
Because there are at least 16 types of circuits that all do the same thing, but using different parts as the years went on. The mechanics never changed, so by removing all the original res/cap and variable res components with something "standardized", I can do it on any of the 16+ types.
Check out the photo I posted earlier, the circuits on the left are about the size of the space.
This analog replacement approach seems more doable, cheaper and less risky to my brain. However, the programmable board in the future is a definite possibility depending on how well this holds up to real world use.
Ain't nothing wrong with the 555 approach. There's good reasons they are still readily available.
FWIW In terms of sheer volume, I don't think the Arduino Pro Mini can be beat http://www.sparkfun.com/products/9220 . It's a bit more expensive at $19, but since it's SMT it can be much lower profile.
B
FWIW In terms of sheer volume, I don't think the Arduino Pro Mini can be beat http://www.sparkfun.com/products/9220 . It's a bit more expensive at $19, but since it's SMT it can be much lower profile.
B
Now if I want to get a 555 which will run on BOTH 3v and 4.5v, do I need 2 different kinds or does a lower power one (keyword?) suffice for both?
And the drain on the system to keep the magnet which I want to power on is 26mA...
And the drain on the system to keep the magnet which I want to power on is 26mA...
Last edited:
This is one where you need to look at the datasheet for the actual part(s) you are considering since the 555 comes in countless varieties.
There should be a specified range for Vcc, so just make sure that both 3.3 and 4.5V are in the range.
IIRC there may be a voltage drop from Vcc to the output of the 555, so you need to make sure that given your supply voltage you have enough oomph on the output to drive your magnet.
Or, as I mentioned earlier you may need to provide a separate output buffer or driver circuit. I know I've driven speakers and other loads directly off of a 555, so the ones I was using had some current capacity.
B
Apparently the LM555 can drive 200mA directly.
This page contains TONS of good info on the 555 chip:
http://home.cogeco.ca/~rpaisley4/LM555.html
R/C calculators, specs, how to drive higher current loads, etc. Should be very helpful.
That is a great, if not better, option for size-constrained projects. I personally don't give it much thought, though. For some reason I don't often wander down the 3.3V MCU path.
I understand your sentiment, but I'm the opposite. I come from the software side of the fence, so thinking of everything in variables, conditionals and loops makes more sense than trying to figure out how much resistance needs to be added to a circuit for a capacitor to determine a 555's frequency. Maybe one day I'll change my religion, but I'm more productive this way.
Also, wiring up a whole bunch of components, switching them out because something didn't work right, recalculating, etc, is less doable and FAR more risky to my brain. If I want to crank something out real quick, I prototype on the MCUs. Then, if I want something (a lot) cheaper, I would then move it to the analog world. Well, probably not, but you get my point. Again with the religion.
We each have our way, based on our strengths and interests. That's what is so great about things like this. More than one way to skin a cat, and we just might learn something from the other cat skinner. Whatever route you take, I wish you good luck! Keep us posted, and maybe even post some schematics once you have it working.
I'll stop hijacking your thread now...
CAREFUL: Read the datasheet. The up to 200mA is valid only for Vcc=15V.
And the chart on the datasheet for output source current is valid only for Vcc=5V-15V.
FWIW there's a 5V version of that too. http://www.sparkfun.com/products/9218
B