12V - 7.29A ac/dc converter project Need Help

SgtWookie

Joined Jul 17, 2007
22,230
Now that I've looked at the datasheet again, I remember why I didn't use the 723's I had for a supply project; the line and load voltage regulation specifications are excellent, but the current limiting function over temperature is very poor. One could do better using a resistor.

So, I'm not going to waste either of our time by pursuing that approach any longer.

The LM317 approach is about as inexpensive, reliable and simple as one could ask for. One IC, and one resistor (or a couple, if you can't find the exact one you need).
 

Thread Starter

Butterworth

Joined May 6, 2009
135
Now that I've looked at the datasheet again, I remember why I didn't use the 723's I had for a supply project; the line and load voltage regulation specifications are excellent, but the current limiting function over temperature is very poor. One could do better using a resistor.

So, I'm not going to waste either of our time by pursuing that approach any longer.

The LM317 approach is about as inexpensive, reliable and simple as one could ask for. One IC, and one resistor (or a couple, if you can't find the exact one you need).
Ok, it makes sense, what you have propsed is to take the lm317 and add the resistor from the adj pin to the input pin? I will pick up the regulator and the resistors tomorrow.
 

Thread Starter

Butterworth

Joined May 6, 2009
135
The input cap is not required, but it is recommended.
Alright, sorry for the novice question here but the Sch. shows the cap as non-polarized, this is correct?
Also is this cap electrolytic or someting else?
I do not want to pick up the wrong cap when I go get the components. Thanks.
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Alright, sorry for the novice question here but the Sch. shows the cap as non-polarized, this is correct?
Also is this cap electrolytic or something else?
I do not want to pick up the wrong cap when I go get the components. Thanks.
Don't worry about it seeming to be a "novice" question. Better to ask and save yourself some trouble later.

You usually want non-polarized caps for such small values. Metal poly film or ceramic are fine; you could even use a "green cap". The small values give you a hint that they are for suppressing oscillations at higher frequencies.

When you start getting over ~0.47uF/470nF, metal poly film and ceramic caps start getting large and expensive - although you can purchase suprisingly large value SMT/SMD caps in 0805 and 0603 sizes (quite small).

Electrolytic capacitors are most often polarized, although you can buy non-polarized electrolytics. The latter are commonly used as motor start caps for AC motors.
 

Thread Starter

Butterworth

Joined May 6, 2009
135
Don't worry about it seeming to be a "novice" question. Better to ask and save yourself some trouble later.

You usually want non-polarized caps for such small values. Metal poly film or ceramic are fine; you could even use a "green cap". The small values give you a hint that they are for suppressing oscillations at higher frequencies.

When you start getting over ~0.47uF/470nF, metal poly film and ceramic caps start getting large and expensive - although you can purchase suprisingly large value SMT/SMD caps in 0805 and 0603 sizes (quite small).

Electrolytic capacitors are most often polarized, although you can buy non-polarized electrolytics. The latter are commonly used as motor start caps for AC motors.
Awesome, I got a small looking cap that is rated to 50V, 0.1uF, should be good ;)

Ok so I got my items to make the CC tester for my leds. I could not find the exact resistor values you stated, but I got 3 different ones to add up to 12.5 in the end. they were all 1W, 5% CF. The only ones they had. ;)

I am going to put this together later on tonight after work, maybe make it on my breadboard first before I solder things together!

After I test each led, I separate them according to their Vf's? What do I do once they are categorized? Do I keep the Vf's the same on the strings of leds when I wire them up to my board?
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
The idea is to make the total Vf of every string come out equal to all of the others, or as near as possible.

Most of the LEDs' Vf will fall within a very narrow range; that will be your "typical" value.

Then you can match up the high Vf's with the low Vf's, so that the total is the same as for all the strings.
 

SgtWookie

Joined Jul 17, 2007
22,230
It's not that bad.

I had 240 blue LEDs on tape. I tested them all, and numbered them as I went through.

The results were recorded in a Microsoft Excel spreadsheet; the sequence number and the Vf.

After the testing/recording was finished, I sorted the list by the Vf in ascending order.

Then I moved the lower half (120 items) of the data up and to the right of the top half.
Then I sorted the 2nd half by Vf in descending order, which pretty well evened up all of the "odd ducks"
 

Thread Starter

Butterworth

Joined May 6, 2009
135
That is a very good way to do it. I will adopt your methodology for this task!

I am going to do all this tonight, then report my findings... I think this led project will have to be pulsed instead of steady current fed? I think Adjuster hinted at that in one of his replies, but may be wrong on this. So that would be the next task!
 

Thread Starter

Butterworth

Joined May 6, 2009
135
Got the results of the voltages.

There are:
68 x 1.42v
7 x 1.43v
4 x 1.41v
1 x 1.44v

Next step is to lay them out, but I need to know if pulsing them is more effective, or use uninterupted current?
 

SgtWookie

Joined Jul 17, 2007
22,230
It's been so long that I forgot what you wanted to do in the first place.

I thought you originally wanted to operate the strings @ 90mA each.

Then when you found out the current would be much reduced, we were talking 100mA per string - something like 13 strings.

Now, do you want them to be dimmable?

That's certainly do-able. You start off setting up the strings so that when they are on 100% of the time that they get 100mA current flow.

Then you add a PWM circuit which is adjustable from nearly 100% to nearly 0%, and switch the current using a power MOSFET.

Is that what you want to do?
 

Thread Starter

Butterworth

Joined May 6, 2009
135
It's been so long that I forgot what you wanted to do in the first place.

I thought you originally wanted to operate the strings @ 90mA each.

Then when you found out the current would be much reduced, we were talking 100mA per string - something like 13 strings.

Now, do you want them to be dimmable?

That's certainly do-able. You start off setting up the strings so that when they are on 100% of the time that they get 100mA current flow.

Then you add a PWM circuit which is adjustable from nearly 100% to nearly 0%, and switch the current using a power MOSFET.

Is that what you want to do?
I need the circuit to be dimmable, have an indicating power led (5mm 2.2Vf@10mA) and amain power switch.

if possible, can a dtdp switch be used to set tje power from low / off / hi ? I can reuse a switch from an old blow dryer if that can be done?
 

SgtWookie

Joined Jul 17, 2007
22,230
I need the circuit to be dimmable, have an indicating power led (5mm 2.2Vf@10mA) and a main power switch.
OK, but how dimmable? Just off/low/high? Or do you want it to be continuously variable?

A power LED would be on all the time that the supply was plugged in. It wouldn't take much power to run it though.

if possible, can a dtdp switch be used to set the power from low / off / hi ? I can reuse a switch from an old blow dryer if that can be done?
That could possibly be done. You will need to figure out what the connections are to the switch.
 

Audioguru

Joined Dec 20, 2007
11,248
Isn't 100mA the max peak allowed current spec?

Usually the peak allowed current also has a very short duration spec.
My MV8191 5mm red LEDs are allowed 200mA max peak current for 100us then an off time of 900us before another peak current pulse.
Their max allowed continuous current is 40mA at room temperature but their recommended operating current is only 20mA.
 

Thread Starter

Butterworth

Joined May 6, 2009
135
OK, but how dimmable? Just off/low/high? Or do you want it to be continuously variable?
The switch only has Hi, Low and Off at the center position only, so I think if applying close to 100% power, say 90% that should give us approx. 90mA? But making it around 50mA at low power? Just for an example...

A power LED would be on all the time that the supply was plugged in. It wouldn't take much power to run it though.
Thats what I need, to indicate power is applied to the unit, because it is IR leds, you can't see it, and I don't need people staring into it to see if it is on. ;)

That could possibly be done. You will need to figure out what the connections are to the switch.
That is no problem. Its a rocker switch with 3 contacts on the bottom side. (Low / Off / Hi) In order of bottom to top.

Isn't 100mA the max peak allowed current spec?

Usually the peak allowed current also has a very short duration spec.
My MV8191 5mm red LEDs are allowed 200mA max peak current for 100us then an off time of 900us before another peak current pulse.
Their max allowed continuous current is 40mA at room temperature but their recommended operating current is only 20mA.
The ones I am using are IR and they are std from what I can tell. I would like to pulse them if running them at 90mA could potentially "fry" them. But if they do fry, in the end, it only cost about $6.00 for 80 leds.
 

SgtWookie

Joined Jul 17, 2007
22,230
Wait a minute - in the very first post of this thread, you gave a Vf @ 100mA - where did you get that 100mA specification from? And is that 100mA continuous, or 100mA for a very short duration?

Show us where you bought them; or better yet, supply a manufacturer's datasheet for the LEDs.
 
Top