PWM Troubleshooting - Need Input

I wired a PWM according to CD's schematics like many brewers here have done. I tested it on a light bulb and can not get it to dim the light. The SSR will have the 115v line OFF when the pot is cranked to 0%, and then passes current to the bulb when the pot reaches around 50%. It does not dim.

I'm wondering if this might be an issue with the timer chip (lm555)? Baring the possibility that I wired this circuit wrong (wrong pins on the chip, i.e.) I wanted to see if anyone has experienced this and if there was an obvious error I made.

Though, I guess this begs the question: Is the PWM supposed to dim the light? Or only switch it on/off?

When the power is left on (50%+) the light does not switch on/off, just stays on.

Here are some pics:

Set up to test... a little cluttered.


Pot turned to ~50%. Led lights on circuit and SSR passes current to bulb.

lustrum

The light should not dim. It should go on/off as per "duty cycle" of your circuit.

Without seeing the schematic it sounds like the cap that sets your timing is not functioning correctly. It sounds like it isn't doing it's charge/discharge cycle to turn the output to the light off and on. You should see the led on the SSR pulse as you adjust the pot as well as the light go on and off with the pulse. Disconnect the SSR and just watch the output pin 3 with the multimeter and see what it does to rule out anything on the SSR's end. You should see the voltage cycle full on and off to what your input is.

As mentioned the light shouldn't dim, just switch on and off. For example, if you have a 2 second cycle and set the pot to 50%, it should switch on for 1 sec, and then off for 1 sec, and if you turn it down to 25% it should switch on for 0.5 sec, and then off for 1.5 sec. What size capacitor did you use? A 4.7uF cap gives a cycle of about 1.5 seconds, and the larger the cap the longer the cycle. Could it be that you have built a circuit with a long cycle time and just aren't waiting long enough to see it work properly?

interplexr said:

Disconnect the SSR and just watch the output pin 3 with the multimeter and see what it does to rule out anything on the SSR's end.

Click to expand...

Thanks I'll try that again and record what it does.

JuanMoore said:

What size capacitor did you use? A 4.7uF cap gives a cycle of about 1.5 seconds, and the larger the cap the longer the cycle. Could it be that you have built a circuit with a long cycle time and just aren't waiting long enough to see it work properly?

Click to expand...

I am using a 4.7uF bipolar electrolytic cap. My understanding is that these can be soldered in any direction. Perhaps I will swap it out with a standard electrolytic cap and see if this makes a difference.

Originally I had wondered if my timer chip was shot, since it was not pulsing.

Thanks for the help so far.

Put a radial lead electrolytic cap (4.7uF) in place of the axial cap and still have the same issue. Also put a meter to the power output and it had no fluctuation, only a solid 5.75VDC once the pot was turned to ~50% and higher.

I'm thinking it is the chip now... I'll swap it out when I get home and see.

What voltage are you sourcing the circuit with if you're only getting 5.75V out?

Cell phone wall wart... 9vdc. The SSR only needs 3vdc to operate.

What resistors do you have in the circuit? Here is a calculator to determine how fast it should be cycling. http://www.electronicdesignworks.com/utilities/555_frequency_calculator/555_frequency_calculator.htm If you are cycling too fast it will give you wierd results like this, since SSRs cannot be used to dim AC current.

lustrum said:

Put a radial lead electrolytic cap (4.7uF) in place of the axial cap and still have the same issue. Also put a meter to the power output and it had no fluctuation, only a solid 5.75VDC once the pot was turned to ~50% and higher.

I'm thinking it is the chip now... I'll swap it out when I get home and see.

Click to expand...

Don't trust your meter to measure the output of a PWM. You may need a scope to see the changes.

The reason is, my meter reads the voltage for a moment before it displays on the screen. When I put it on my PWM, it says 0v until I get upwards of 40%. Then it says 9v the whole time - It's simply because the response time of the meter is higher than the off cycle of the PWM at that setting. - IOW, it's not off long enough for my digital meter to display 0v before the PWM fires high again.

Hook a 9v lamp, or LED (With the proper voltage divider) - They should be able to show you that current is flowing, but still have the response time to let you see the "off" portion of the pulse.

chunk1227 said:

What resistors do you have in the circuit? Here is a calculator to determine how fast it should be cycling. http://www.electronicdesignworks.com/utilities/555_frequency_calculator/555_frequency_calculator.htm If you are cycling too fast it will give you wierd results like this, since SSRs cannot be used to dim AC current.

Click to expand...

I have two 470k resistors and a 1k per CD's schematic (note: mind has one less b/c i do not use the "power" led): http://home.highertech.net/~cdp/boilnew/boilnew.htm

SweetSounds said:

Don't trust your meter to measure the output of a PWM. You may need a scope to see the changes.

The reason is, my meter reads the voltage for a moment before it displays on the screen. When I put it on my PWM, it says 0v until I get upwards of 40%. Then it says 9v the whole time - It's simply because the response time of the meter is higher than the off cycle of the PWM at that setting. - IOW, it's not off long enough for my digital meter to display 0v before the PWM fires high again.

Hook a 9v lamp, or LED (With the proper voltage divider) - They should be able to show you that current is flowing, but still have the response time to let you see the "off" portion of the pulse.

Click to expand...

SweetSounds: i think it might be working at this point. I tested it on a 115v line powering a light bulb and and changed out the 50v 4.7uF cap to a 100v 4.7uF cap. Now it pulses once the pot is turned at 50% or higher... I can see the bulb go off...on...off...on. Funny how it does nothing when the pot is turned below 50%... is this common to anyone else's experiences?

Thanks

lustrum said:

I have two 470k resistors and a 1k per CD's schematic (note: mind has one less b/c i do not use the "power" led): http://home.highertech.net/~cdp/boilnew/boilnew.htm



SweetSounds: i think it might be working at this point. I tested it on a 115v line powering a light bulb and and changed out the 50v 4.7uF cap to a 100v 4.7uF cap. Now it pulses once the pot is turned at 50% or higher... I can see the bulb go off...on...off...on. Funny how it does nothing when the pot is turned below 50%... is this common to anyone else's experiences?

Thanks

Click to expand...

That still doesn't sound right - A 50v 4.7uF cap and a 100v 4.7uF cap are the same thing - Just with different ratings - They both have a capacitacne of 4.7 micro farads.

You should see a change in on vs off time as you sweep through the dial. It may not be linear, but at 25% it should still come on for somewhere around 1/2 of a second. (assuming your interval is 2 seconds)

Right. That's what I was thinking, but can only guess that the cap was bad. There must still be some flaw in this PWM. Do they typically show a pulse at low pot %? My thought was that it would show a very slow pulse at low % and a fast, almost undetectable pulse at high %.

lustrum said:

Right. That's what I was thinking, but can only guess that the cap was bad. There must still be some flaw in this PWM. Do they typically show a pulse at low pot %? My thought was that it would show a very slow pulse at low % and a fast, almost undetectable pulse at high %.

Click to expand...

The cycle should be the same length no matter what the pot is set at, somewhere around 1.5-2 seconds long. The difference should be the ratio of on to off in that cycle. For example, at 25% with a 2 second cycle it should be on for 0.5 sec, and then off for 1.5 sec. At 50% it should be on 1 sec and off 1 sec. At 75% it should be on 1.5 sec and off 0.5 sec.

Understood. But at 25% it does not come on at all (as far as I can see). The bulb never flashes. At 50% it cycles on then off in 1 second intervals.

Is this how it should function?

If you're only getting 5V out, that seems to be a problem to me. You should have 9V. I had a wiring issue and burned up an IC and had funny output voltages that did weird things with the pot. With a 4.7uF cap and the 1K resistor you should have about a 3 second period so at 50% on the pot you'd have 1.5 seconds on and 1.5 second off. If the 555 IC is hot that's a sure sign something is wrong. If the IC is mounted in a holder you can pull it and check the terminals for the proper voltages to check wiring issues if you haven't already.

^^ +1
I always use a DIP socket so that soldering the connections doesn't fry the IC. Hard to tell if one was used from the pics. Either way I'd try replacing the IC and see what happens.

Good points, gents. I think I will redo this circuit on a freshly cut PCB. I also like the idea of using a DIP socket for the IC.