My First Pulse Width Modulator Mod

[CraigHB]

Hmm, sounds good. I may have to start looking for a good hotplate (don't have one). I have a nice temperature probe on my DMM so I can use that to set up a proper soldering profile on it. The aluminum plate as a heat spreader sounds like a big helper.

Soldering each component by hand is quite tedious so even if I can use a hotplate to solder only a subset of components, that will be a real time saver.

 

[bstedh]

Hmm, sounds good. I may have to start looking for a good hotplate (don't have one). I have a nice temperature probe on my DMM so I can use that to set up a proper soldering profile on it. The aluminum plate as a heat spreader sounds like a big helper.

Soldering each component by hand is quite tedious so even if I can use a hotplate to solder only a subset of components, that will be a real time saver.

I said hotplate because that is what they said over at hackaday but it is actually an electric skillet. Either or would work the same. One guy was even using a mug warmer to pre heat the board prior to hand soldering. I have a little hot pot water boiler that I am going to try and use as it heats up very quickly to the selected temp and I can use the adjust on the side to run my temp curve fairly accurately once I figure out its temperature range. At least that's my plan =]

If it works good, making a pwm in a small form factor with smd components may get a lot more practical for more people.

Also if you find usable components on an old mother board they showed that is is real simple to remove them by using a heat gun on the back side of the board and then just shaking the component off or slightly prying it. I am going to go over my stack of motherboards to see what I can pull out them. I also have some computer power supplies that I am also going to rip into to see if there are usable pieces and parts. The wife will be happy to see this pile of stuff reduced to a smaller pile =] And hopefully I can reduce my digi key order...

 

[CraigHB]

A couple years ago, I flushed all my old computer parts. I had built up quite a collection of old stuff. Filled a couple large boxes. The only things I slavaged off the old motherboards were a few small heatsinks and the coin cell holders. The coin cell holders come in real handy for other projects and it's hard to find those nice small ones they use on motherboards.

Years and years ago, I used to actually break stuff down and save the parts. That was before before the internet and SMD components. Nowadays, I can get SMD resistors online for a few cents each. MLC caps are super cheap too. I've actually abandoned the use of electrolytic caps for the most part since the higher value MLC caps are becoming competetive in price. For the real high value caps, I've been using the tantalum SMD ones, much more compact. Chips can be expensive, but I try to order them online via free samples. I actually have quite a collection of parts at this point, but few of them salvaged. It's great to be doing this stuff nowadays with the convenience of the internet. When I first started doing this stuff, I was always making trips to the local electronics shop and spending way too much for components.

Anyway, I'll start looking around for some kind of "plate" style heater. Sure I can find something good and inexpensive online.

 

[clutchjunkie]

Here is the circuit I am currently using, sorry it took me so long but real life kept getting in my way, lol. I laid it out like this to try to keep it simple for the non-EE types, it is pretty much how you would put it together on a pre-drilled board. I am also lazy and this was quick.

C1 is a .01uf capacitor (it has the number 103 on it)
R1 is 220 ohm 1/8 watt
R2 is 150 ohm 1/8 watt

I might to play around with the potentiometer a bit and go up to 300 ohm or maybe 500 ohm for more adjustability, but it is not necessary. This circuit works really well for dual coils and LR atties and cartos, I have been using it for close to a month now.

 

[asnider123]

Here is the circuit I am currently using, sorry it took me so long but real life kept getting in my way, lol. I laid it out like this to try to keep it simple for the non-EE types, it is pretty much how you would put it together on a pre-drilled board. I am also lazy and this was quick.

C1 is a .01uf capacitor (it has the number 103 on it)
R1 is 220 ohm 1/8 watt
R2 is 150 ohm 1/8 watt

I might to play around with the potentiometer a bit and go up to 300 ohm or maybe 500 ohm for more adjustability, but it is not necessary. This circuit works really well for dual coils and LR atties and cartos, I have been using it for close to a month now.

Bless you, Clutch! I had put the PWM on the back burner, it just had too many components and was too hard to stuff into the box. Your design is clean and simple and requires the minimum of components. Maybe I can now get back to that project You don't mind if I use your basic design and maybe play with the resistors a bit? Thanks, Alan

 

[asnider123]

Here is the circuit I am currently using, sorry it took me so long but real life kept getting in my way, lol. I laid it out like this to try to keep it simple for the non-EE types, it is pretty much how you would put it together on a pre-drilled board. I am also lazy and this was quick.

C1 is a .01uf capacitor (it has the number 103 on it)
R1 is 220 ohm 1/8 watt
R2 is 150 ohm 1/8 watt

I might to play around with the potentiometer a bit and go up to 300 ohm or maybe 500 ohm for more adjustability, but it is not necessary. This circuit works really well for dual coils and LR atties and cartos, I have been using it for close to a month now.

In looking at your schematic, have a question: is pin 8 tied to the right side of the atty coil? That's the only way I can see that the atty could get a positive connection

Thanks

 

[clutchjunkie]

Yes, pin 8 and the positive side of the atty are connected. If you wanted to use a tactile switch in this, you could also tie the atty + side straight to the main on/off switch and run the tactile switch between that connection and pin 8. If you do that though you might need a 22k ohm resistor between the gate and source of the mosfet to make sure it shuts off, but you would have to breadboard that to see if the resistor is necessary.

I just use a big 5 amp pushbutton to keep it simple.

 

[asnider123]

Yes, pin 8 and the positive side of the atty are connected. If you wanted to use a tactile switch in this, you could also tie the atty + side straight to the main on/off switch and run the tactile switch between that connection and pin 8. If you do that though you might need a 22k ohm resistor between the gate and source of the mosfet to make sure it shuts off, but you would have to breadboard that to see if the resistor is necessary.

I just use a big 5 amp pushbutton to keep it simple.

Hi Clutch, well I breadboarded it exactly as you showed in your diagram, substituting a 500ohm pot cause I don't have a 200 .. the FET latches and stays latched. Suspect the 555 not oscillating but not sure because I have no scope. Drain reads same as source: -8.21v

Tried replacing fet, replaced 555, replaced resistors, no change. Reversed diode, drain drops to 0v and stays there.

One other thing .. only using LED for load, does the circuit use the resistance of the load as part of the oscillator? Perhaps that's the prob ???

Any thoughts please ???????

 

[clutchjunkie]

No matter what I have my pot set at, all of my voltmeters show full battery voltage at the atty connection. I actually had to use a carto when I was building the circuit to get the resistor values right, dual coils are great for that because they can handle a lot of voltage.

 

[asnider123]

No matter what I have my pot set at, all of my voltmeters show full battery voltage at the atty connection. I actually had to use a carto when I was building the circuit to get the resistor values right, dual coils are great for that because they can handle a lot of voltage.

So with a 3.2 ohm 801 carto on it (what I have on hand), I should begin getting voltage readings across the atty? I have a True RMS voltmeter, it should let me know the voltage drop across the carto, shouldn't it? Thanks for your help.
Alan

 

[asnider123]

No matter what I have my pot set at, all of my voltmeters show full battery voltage at the atty connection. I actually had to use a carto when I was building the circuit to get the resistor values right, dual coils are great for that because they can handle a lot of voltage.

OK, put a 3.2ohm carto on as a load .. smokes like a ........ experimented with the pots I had on hand ... 100 ohms, 1K, 10k ... at 10K it has the range I was shooting for (about 3 low to 5+ high), I chickened out before I found the high, my old clearo is pretty dry and I didn't want to fry it. I got these readings across the carto. With no carto: source and drain voltage both at batt voltage (about 8.2v)

Put a filled CE2 on it and hit it ... WOW!!! Cough-cough .... one of best hits I ever had .. carto gets pretty hot, I don't trust the voltage reading (4.98) .. perhaps it's correct. Well, let's put this baby together and see if it melts the box or carto
Alan





Edit: I checked the components and, though the carto was really cooking, the FET and 555 stayed really cool Guess it is working correctly. Cool!

 

[clutchjunkie]

Actually, I have no idea. I build them "to taste ", the only other indicator I use is how bright the led is before I connect the atty or carto. You should be able to tell the difference in brightness from one end of the pot's travel to the other. I use a purple led with a 470 ohm resistor as a load.

A 3.1ohm atty should be a nice "safe" one to test with since they can usually handle 5v or more. Start with the pot adjusted all the way to the pin 8 resistor as this is the lowest duty cycle.

Edit : dah!! You beat me to the post. Yes, cartos will run a little hot on this setup.

 

[Java_Az]

Just got back from radio shack and more importantly the beer store. Going to Blow the dust off the breadboard and check this circuit out, hopefully without burning down the house. Going to be interesting to compare it to The Ti switching reg i been using. If i like it, I will see just how small i can make it on a circuit board with smd resistors and caps super sot6 fet looks like 8 sop is the smallest pin package for a timer.

Thanks for the thread asnider and Thanks to all those who have contributed to it ......Cheers

 

[bstedh]

Ok here is a board design for this circuit. However I want to add a zener cut off to pin 8 and and indicator LED.

The board is only 1/2" X 3/4". or 21mm X 12mm

 

[bstedh]

What would be a good cut out voltage for the zener? I was thinking about 6.4V but don't know how much battery sag would come into play on this and if I should go lower than that.

 

[Java_Az]

Ok i added a 0603 resistor off of pin 3 in case there is a latching problem you can tie it to the low side. I went ahead and made a foorprint for the Ti 555 chip from scratch. It now shows the body of the chip the legs will hit about half way down on the visible pads. going to wait to hear back from you on the where the pot needs to be before I move it.

switch side

Part list:

1 10 RNCP0603FTD150RCT-ND RES 150 OHM 1% 0603 SMD
2 10 RHM220DCT-ND RES 220 OHM 1/5W 5% 0603 SMD
3 10 296-6502-1-ND IC OSC MONO TIMING 500KHZ 8-SOP Changed to 296-16792-ND IC OSC MONO TIMING 500KHZ 8-SOIC
4 10 3352W-501LF-ND POT 500 OHM THUMBWHEEL CERM ST
5 10 641-1011-1-ND DIODE SCHOTTKY 30V 100MA 1005 maybe change to this http://search.digikey.com/scripts/dksearch/dksus.dll?pname&site=us&lang=en&WT.z_cat_cid=Dxn_US_US2011_Catlink&name=641-1283-1-ND
6 10 490-1755-1-ND CAP CER 11000PF 50V 5% C0G 1206 Changed to 445-5099-1-ND CAP CER 10000PF 25V X7R 0603
7 10
NCNR DMN2020LSN-7DIDKR-ND MOSFET N-CH 20V 6.9A SC59
8 10 P12301SCT-ND SWITCH LT 4.7MMX3.5MM 350GF SMD
9 10 BZX84C6V2WDICT-ND DIODE ZENER 6.2V 200MW SC70-3 Digi-Key has discontinued this item; limited quantity available.

10 10 0805 LED ???????????
11 10 0603 resistor for LED ??
12 ?? 0603 Optional Resistor for gate ???????????

 

[bstedh]

Trying to find a dremel disk to cut out a proto board... I really need to organize a little better...

That zener was just a quick search. I couldn't find one in the voltage I wanted of 6.4...

 

[bstedh]

Pot can be moved for sure there is quite a bit of room to play with. Also need to know what side of the board you want it on. I just guessed and though it would go opposite side of the switch. Just need to know that and how much clearance you from the end it needs.

That foot print for a 8-soic ( 555timer) was already in the program. Real life it doesnt look like that you will have pins to solder. I checked it to make sure the pads were in the right place but i didnt check the width of the pads they are a bit skinny. I will go ahead and make a footprint of it and fix it up with a proper silk of how the chip really looks.

I went ahead and bread boarded this circuit minus the switch and zener . It works for me but i only have a 250ohm. My next lowest is a 10k pot seems to work better on the 10k for me anyways. Also might need to add a resistor to the pin3 gate trace to pull it low. From the other thread if i remember right there was a latching problem. Also i believe pin 5 should be capped to ground if not used( not sure if it even matters or not ).

I might mess with it a bit tonight and throw it on the PWM thread where the circuit comes from.

I plan on ordering a couple different POT's. My initial plan was to glue a lever on the POT to protrude from the case but I don't think I would be able to get the range needed doing that. So, I may just work on cutting a slot for the edge to stick out just enough to get a fingernail on it.

I was also thinking of a slider POT but couldn't find one that was small enough.

 

[bstedh]

Ok, I did some dry fitting and we will need to move the POT as far back from the edge as possible and also move the output of the fet to that edge so the wire does not interfere with the POT. It can be done by running the wire on the switch side if it is too much of a pain however.

Here is how the board will sit inside the tube. The drawing is no where to scale and I may need to use a spacer on the POT to get it to protrude properly from the body. I can't do much about that as the switch has to be close enough to the top to allow the button to push it. 1/2" X 3/4" looks like it will be about perfect.

 

[bstedh]

This may be a closer size representation. I may look into the lever idea. I just have to experiment with the POT size to get about 3.5V to 6V from a 1/4 turn of the pot. Then the pot just needs to be set properly before gluing the lever on.

The board itself will be backed by a spacer to keep it in place.