Author Topic: 555 timer PWM voltage drop  (Read 3678 times)

0 Members and 1 Guest are viewing this topic.

Offline rodrigopires

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pt
555 timer PWM voltage drop
« on: May 09, 2016, 07:23:20 pm »
Hi everyone! I have recently made a 555 timer pwm for my hho dry cell. It is supposed to deliver 12v, but when i connect it to the cell, on full power, the voltage drops to around 10V and the cell does not go to full potencial. Without the pwm, at 12v the cell consumes around 43A, but with the pwm because the voltage drops, it only goes to around 24A, and i wanted it to go from 0 to the full 43A. Is there something i can do in the pwm circuit to decrece that voltage drop? Higher frequency maybe? Btw, the main PS is a hp ps-3701-1, 12V 60A.

R1-1KOhm ; R2-10Ohm ; P1-50KOhm ; C1-100nF ; C2-10nF ; D1,2-1N4148 ; Q1-IRFP064N
 

Offline Andy Watson

  • Super Contributor
  • ***
  • Posts: 1732
Re: 555 timer PWM voltage drop
« Reply #1 on: May 09, 2016, 07:39:01 pm »
The +5V available from the 555 and TC4426 is barely enough to turn the FET on. Figure 1. on the IFRP064 datasheet suggests that you will need at least 5.5V gate voltage to attain 43A - I would go higher (8-10V) to ensure that the FET is not operating in its linear region.
 

Offline rodrigopires

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pt
Re: 555 timer PWM voltage drop
« Reply #2 on: May 09, 2016, 07:53:12 pm »
It says on the Vgs 2V min to 4V max, and it's a IRFP064N, different from the IRFP064. (http://www.hestore.hu/files/irfp064n.pdf)
 

Offline Andy Watson

  • Super Contributor
  • ***
  • Posts: 1732
Re: 555 timer PWM voltage drop
« Reply #3 on: May 09, 2016, 08:22:53 pm »
It says on the Vgs 2V min to 4V max, and it's a IRFP064N, different from the IRFP064. (http://www.hestore.hu/files/irfp064n.pdf)
It says 2 to 4V gate threshold, that is the voltage at which the FET begins to conduct - and the "Conditions" are defined as \$I_D = 250\mu A\$.
 

Offline rodrigopires

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pt
Re: 555 timer PWM voltage drop
« Reply #4 on: May 09, 2016, 08:32:35 pm »
so if i use lets say 8V on the 5V input, going 8V to the gate, that would improve the voltage drop and efficiency on the pwm? How about the frequency of the pwm? Does that affect the total power output?
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 1927
  • Country: us
Re: 555 timer PWM voltage drop
« Reply #5 on: May 09, 2016, 09:00:54 pm »
Why are you only using 5V to power the 555 and the TC4426? Just go ahead and supply these with the same 12 volts as you are using for the hydrolysis cell. The mosfet is likely stuck in the linear range with only 5 volts (or a bit less really) on the gate and is wasting power as heat and isn't fully turning on.

The easiest person to fool is yourself. -- Richard Feynman
 

Offline rodrigopires

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pt
Re: 555 timer PWM voltage drop
« Reply #6 on: May 09, 2016, 09:15:48 pm »
Originally i had it all on 12v, but while testing some things weren't right, like i had voltage on the drain with the pwm circuit off, and some parts burned (like the mosfet and driver chips) so i isolated the pwm to a 5v rail and the mosfet source to 12v and it works fine, just that problem with the power output. The 555 is rated at 18V max input, the  TC4426 at 18V so i dont see  problem with the 12v pwm. But will the gate take 12V?
 

Offline Andy Watson

  • Super Contributor
  • ***
  • Posts: 1732
Re: 555 timer PWM voltage drop
« Reply #7 on: May 09, 2016, 09:23:52 pm »
But will the gate take 12V?
The gate will withstand \$\pm\$20V, but absolutely no more! So it should be ok to drive the whole circuit from 12V.  The PWM frequency is relatively independent of supply voltage (the 555 trips at 1/3 and 2/3 Vcc - what ever that is).
Other than the hho cell, is there something else in the drain circuit to limit the current?
 

Offline rodrigopires

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pt
Re: 555 timer PWM voltage drop
« Reply #8 on: May 09, 2016, 09:36:17 pm »
I will try the 12V see if it improves. Not at all time, just a 5 mOhm shunt to measure the current.
« Last Edit: May 09, 2016, 09:41:10 pm by rodrigopires »
 

Offline Audioguru

  • Super Contributor
  • ***
  • Posts: 1508
  • Country: ca
Re: 555 timer PWM voltage drop
« Reply #9 on: May 09, 2016, 10:20:39 pm »
The Mosfet conducts only 0.25mA when its gate to source is its 2V to 4V threshold voltage. The graph fig. 1 is for a "typical" one that you cannot buy, you might get a minimum one. The written spec's on the datasheet say that with a 10V gate-source voltage its on-resistance is guaranteed to be very low and it conducts very well.
 

Online Dave

  • Super Contributor
  • ***
  • Posts: 1217
  • Country: si
  • I like to measure things.
Re: 555 timer PWM voltage drop
« Reply #10 on: May 10, 2016, 07:08:55 am »
You're all missing the fact that the circuit that he used can't reach 100% duty cycle, or anywhere near that. To get close to that, C1 would need to charge up practically instantly. Even with the potentiometer turned the whole way to the left, R1 is slowing the charging down.

Remove R1, disconnect pin 7 and instead connect pin 3 to that node, test the circuit, remeasure.
<fellbuendel> it's arduino, you're not supposed to know anything about what you're doing
<fellbuendel> if you knew, you wouldn't be using it
 

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 1927
  • Country: us
Re: 555 timer PWM voltage drop
« Reply #11 on: May 10, 2016, 08:59:36 am »
You're all missing the fact that the circuit that he used can't reach 100% duty cycle, or anywhere near that. To get close to that, C1 would need to charge up practically instantly. Even with the potentiometer turned the whole way to the left, R1 is slowing the charging down.

Remove R1, disconnect pin 7 and instead connect pin 3 to that node, test the circuit, remeasure.

Er.... on my breadboard, using the component values listed on the OP's schematic, the duty cycle ranges from a bit over 2 percent ON to 99.6 percent ON. Since the driver chip is inverting, this translates to slightly less than 98 percent ON at the mosfet.

555 Pin 3 output shown below, with pot turned to either extreme. (I don't have a TC4426 on hand but you get the idea, I hope...)

What am I doing wrong?
« Last Edit: May 10, 2016, 09:15:13 am by alsetalokin4017 »
The easiest person to fool is yourself. -- Richard Feynman
 
The following users thanked this post: rodrigopires

Offline alsetalokin4017

  • Super Contributor
  • ***
  • Posts: 1927
  • Country: us
Re: 555 timer PWM voltage drop
« Reply #12 on: May 10, 2016, 10:02:17 am »
Originally i had it all on 12v, but while testing some things weren't right, like i had voltage on the drain with the pwm circuit off, and some parts burned (like the mosfet and driver chips) so i isolated the pwm to a 5v rail and the mosfet source to 12v and it works fine, just that problem with the power output. The 555 is rated at 18V max input, the  TC4426 at 18V so i dont see  problem with the 12v pwm. But will the gate take 12V?
Yes, the gate will take 12 volts just fine. And since your load is almost purely resistive you shouldn't have problems with inductive spiking. So why your parts burned is a mystery to me, unless your load is shorted and you are pulling the power supply's full 60 amps through the mosfet and it isn't on a heatsink. You may want to put some small, high-power rated resistance in series with your electrolysis cell, or set the current limit (if possible) on the power supply to 43 amps.

" voltage on the drain with the pwm circuit off " I'm not sure exactly what you mean here. With the inverting driver chip, when the input to the 4426 from the 555 is LOW the output to the mosfet Gate is HIGH. SO with no signal from the PWM the mosfet should be turned ON 100 percent of the time. Also, when a mosfet in the low-side switch configuration is OFF the Drain voltage is normally HIGH, and vice versa. ( I'm not sure if this still holds true with an electrolysis cell being the load .... )
The easiest person to fool is yourself. -- Richard Feynman
 

Offline Andy Watson

  • Super Contributor
  • ***
  • Posts: 1732
Re: 555 timer PWM voltage drop
« Reply #13 on: May 10, 2016, 10:16:29 am »
What am I doing wrong?
Nothing. Those are the values that you would calculate for this circuit. With a 1k charging resistor and 50k discharge path you would guesstimate something in the region of 98%  duty cycle.
 

Offline rodrigopires

  • Regular Contributor
  • *
  • Posts: 81
  • Country: pt
Re: 555 timer PWM voltage drop
« Reply #14 on: May 11, 2016, 06:48:13 pm »
So today i tried the 555 and tc4426 on 12V, and it was it! Now i got a control from 0 up to 50A! Exactly what i wanted! It really wasn't good at 5V. Here's a graph with some measurements i took.

Thanks for all the help! That's why i like so much this forum, really great people and  can always find the answer to the problem!
 

Online Dave

  • Super Contributor
  • ***
  • Posts: 1217
  • Country: si
  • I like to measure things.
Re: 555 timer PWM voltage drop
« Reply #15 on: May 11, 2016, 08:06:46 pm »
You're all missing the fact that the circuit that he used can't reach 100% duty cycle, or anywhere near that. To get close to that, C1 would need to charge up practically instantly. Even with the potentiometer turned the whole way to the left, R1 is slowing the charging down.

Remove R1, disconnect pin 7 and instead connect pin 3 to that node, test the circuit, remeasure.

Er.... on my breadboard, using the component values listed on the OP's schematic, the duty cycle ranges from a bit over 2 percent ON to 99.6 percent ON. Since the driver chip is inverting, this translates to slightly less than 98 percent ON at the mosfet.

555 Pin 3 output shown below, with pot turned to either extreme. (I don't have a TC4426 on hand but you get the idea, I hope...)

What am I doing wrong?
Oops. Completely missed the component values written on the bottom of op's message. Didn't look up the driver chip either. :-[
<fellbuendel> it's arduino, you're not supposed to know anything about what you're doing
<fellbuendel> if you knew, you wouldn't be using it
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf