Home     Contact     Forum     Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science     RSS     Terms of services     Privacy policy  
   
 Home     Forum     Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science   

Author Topic: Pcb design for car led lights  (Read 10443 times)

0 Members and 1 Guest are viewing this topic.

cheerio

  • Sr. Member
  • ****
  • Posts: 306

kam

  • Administrator
  • Hero Member
  • *****
  • Posts: 1849
Re: Pcb design for car led lights
« Reply #31 on: March 10, 2014, 20:27:56 PM »
http://www.ebay.de/itm/AideTek-VC97-3999-Auto-Tester-Multimeter-AC-DC-RC-Schiff-wartet-Buzz-EU-/281229192896?pt=Mess_Pr%C3%BCftechnik&hash=item417a90eec0
check this one. it is cheap but it does the job pretty good.
Did you have the chance to test this in comparison with an accurate multimeter???

I'd recommend an Extech multimeter which has a very good price to quality ration, its not considered as a cheap multimeter though since you won't find it for under 50 euros. Its not a fluke, but it has a fifth of its price and a reasonable good accuracy.
« Last Edit: March 10, 2014, 20:33:04 PM by kam »

cheerio

  • Sr. Member
  • ****
  • Posts: 306
Re: Pcb design for car led lights
« Reply #32 on: March 11, 2014, 03:11:06 AM »
i use the aidetek all the time. i could not verify it vs a professional one yet but i did some tests and calculations. precision is pretty good, but you cannot use it in the 0.0x Ohm range. the capacitance meter is not accurate too, the temperature precision is ok. the current precision is good for most circumstances but sux in <1ma ranges

kam

  • Administrator
  • Hero Member
  • *****
  • Posts: 1849
Re: Pcb design for car led lights
« Reply #33 on: March 11, 2014, 08:40:42 AM »
from the day that i got my fluke my life as an ee changed. I was reverse engineering a commercial dimmer once for a report and i was told that this dimmer had a current limiter. The limiter was used to shut down the output if power exceeds 190Watts. So i knew that there had to be a specific resistor of several watts but very low resistance. Typically its a 0.2 to 0.4 ohms 2 watts, but there was NO SUCH THING! And i was quite curious to see how it worked. I then noticed that there were some wire connections on the PCB (jumpers). In the beginning they seemed normal as it is a 1-side PCB. But then i noticed that a specific line that goes from the triac to the AC had these jumper wires going back and forth for no obvious reason. I measured the resistance of the wires, 0.25 Ohms... Problem solved.


cheerio

  • Sr. Member
  • ****
  • Posts: 306
Re: Pcb design for car led lights
« Reply #34 on: March 11, 2014, 21:07:27 PM »
i am sure that a fluke is smth most of us want to have. but those things are expensive :(

filarhos

  • Newbie
  • *
  • Posts: 22
Re: Pcb design for car led lights
« Reply #35 on: April 13, 2014, 17:17:45 PM »
Hi after a short break (exams were on the way) I am back and I will try to make it happen.

I will go to buy some new resistors 3.3 to 2.2 ohm (according to calculations are good enough) and I was wondering if I should buy new transistors and mosfet with better values?
The circuit has changed a little: instead of using only one led I will use two leds in parallel.
So instead of 2volts we will have 4 volts but hte curent will be the same.

The light output is good but the numbers don't agree with the theory. The mosfet still gets hot ~40C.
What to do (except buying new resistors)? Solder all this together or change something (the circuit itself with new types of mosfet)?


« Last Edit: April 13, 2014, 20:19:51 PM by filarhos »

kam

  • Administrator
  • Hero Member
  • *****
  • Posts: 1849
Re: Pcb design for car led lights
« Reply #36 on: April 14, 2014, 23:27:18 PM »
The circuit has changed a little: instead of using only one led I will use two leds in parallel.
So instead of 2volts we will have 4 volts but hte curent will be the same.

This is a series connections. If you connect them in parallel, the voltage s the same, the current is doubled, the heat dissipation is squared. So you should connect them in series, the voltage is doubled, the current is the same, the heat dissipation is the same.

filarhos

  • Newbie
  • *
  • Posts: 22
Re: Pcb design for car led lights
« Reply #37 on: April 15, 2014, 11:41:36 AM »
At 28th post Kam you said that the mosfet works well and the results are reasonable, so I dont have to go and buy another?

The first step is to take some new resistors 2watt 2.2 to 3.3 ohm 1% and then test the results.
About the RG should I stay at the 10k?

P.S. I will go and compare my multimeter with a friends.

Thank you!!! 

kam

  • Administrator
  • Hero Member
  • *****
  • Posts: 1849
Re: Pcb design for car led lights
« Reply #38 on: April 15, 2014, 23:23:40 PM »
For your next tests i think you do not need a new mosfet. And yes, stay at 10K Rg for your tests.
Test your mutlimeter with a fluke.

filarhos

  • Newbie
  • *
  • Posts: 22
Re: Pcb design for car led lights
« Reply #39 on: April 17, 2014, 15:25:01 PM »
I had the idea to check the circuit without the LM2596 and I gave 11.5v knowing that the mosfet will be hot.
The results I recorded was very close to the theoretical (~0.01A). So I tryied to increase with the LM2596 the voltage. At about 7.5v I had the same results as without the LM2596. The mosfet was hot without the Lm2596 but not very hotter (needs a cooler to be sure) than without.
So maybe the 6v was not enough for the cisrcuit because above the 7.5v we have the excpected results. 

For your next tests i think you do not need a new mosfet. And yes, stay at 10K Rg for your tests.
Test your mutlimeter with a fluke.

I went to two stores near and none of them had resistances at 1%.
I will use the ones I have. I will check all of them and I will choose those they are closer to the values I want not the values they should be.
e.g. I want 5.6ohm I have in series 2.2 and 3.3 that measured is 5.8 ohm is close enough to the expected 5.5 ohm.

What do you think?

P.S. About the tolerance is the excpected value it does not change over the time, not dramatically though?


« Last Edit: April 17, 2014, 15:27:36 PM by filarhos »

cheerio

  • Sr. Member
  • ****
  • Posts: 306
Re: Pcb design for car led lights
« Reply #40 on: April 17, 2014, 21:17:46 PM »
just check ebay for small quantities of 1% resistors

kam

  • Administrator
  • Hero Member
  • *****
  • Posts: 1849
Re: Pcb design for car led lights
« Reply #41 on: April 17, 2014, 23:36:29 PM »
I think you will not find 1% at these values/power... 5% is ok. The resistance does not change over time. So, a 3.3Ohm at 5% can be from 3.15 up to 3.45, so if you want to be academically correct then run the power calculations for lowest resistor and current for highest.

filarhos

  • Newbie
  • *
  • Posts: 22
Re: Pcb design for car led lights
« Reply #42 on: April 18, 2014, 14:01:15 PM »
Thank you both!!!

I will do some testings with the resistors to match the outcome with the other led and I will finish with this project.
Photos in a few days.

filarhos

  • Newbie
  • *
  • Posts: 22
Re: Pcb design for car led lights
« Reply #43 on: July 06, 2014, 20:21:34 PM »
The project went great thank you all for all the help you gave me!!!

I will upload a video with the leds running shortly, I believe within this week!!! They look awsome!!!

kam

  • Administrator
  • Hero Member
  • *****
  • Posts: 1849
Re: Pcb design for car led lights
« Reply #44 on: July 12, 2014, 21:59:53 PM »
looking forward to see the video!