| Author |
Message |
|
|
 Posted: Dec 20, 2011 - 01:07 PM |
|
Joined: Dec 15, 2011
Posts: 12
|
|
R13 and R14 are now next to the pins.
How many amps should I make the PTC for the main V+ on the LM2940? |
|
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 01:12 PM |
|
Joined: Dec 15, 2011
Posts: 12
|
|
I was just thinking that if I for a micro with more i/o's, I could duplicate the output circuits and create channels for the rear lights as well since I'm going to be converting the tails to LEDs eventually as well.  |
|
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 03:07 PM |
|
Joined: Nov 02, 2009
Posts: 3239
Location: Zelenograd, Russia
|
|
Now your circuit looks fine.
The fuse trip current should be rated twice the sum of the LED bars current.
Instead of using an MCU with more pins you can just duplicate your MOSFET outputs to the tail lights as well.
Do not forget to connect a current limiting resistor in series of each of your LED bars. Since the LED bar current is high enough (50..100 mA I suppose), you should use the higher wattage series resistors, not the regular 0603/0805 ones. |
|
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 03:36 PM |
|
Joined: May 21, 2003
Posts: 341
Location: Malden MA
|
|
Getting closer. Some of these are stylistic preferences.
- You have two different nodes called V+ (one on the left of F1, one on the right of D1)
- Likewise, you have two VPARKs
- Vcc and Ground symbols are free. Use them instead of wires to simplify your schematic
- I would also use named nets for the connections to the ISP header
- In general, schematics show positive symbols pointing up, ground and negatives down. That makes the input protect circuit easer to read
- Add the zener voltages to the schematic to make it more self-documenting
- Avoid 4-way junctions on schematics (liek the one at the regulator's ground pin)
- Don't draw wires through symbols (see Vcc running thru IC1)
- You have two different symbols for non-polarized caps. Unless there's a good reason, I would stick to one
- You will be blinking some decent loads; I would add some bulk capacitance to the V+ rail
- Your FETs have multiple source and drain terminals. They should all be hooked up.
- R13 an R14 should be connected from gate to ground
- R4 and R5 look a bit large to me (I haven't run the numbers)
/mike |
|
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 03:48 PM |
|
Joined: Nov 02, 2009
Posts: 3239
Location: Zelenograd, Russia
|
|
|
n1ist wrote:
You have two different symbols for non-polarized caps. Unless there's a good reason, I would stick to one
These are just the artefacts of PDF creation.
Quote:
You will be blinking some decent loads; I would add some bulk capacitance to the V+ rail
Bad idea. Guess why.
Quote:
R13 an R14 should be connected from gate to ground
No they shouldn't. Re-read the thread to see why.
Quote:
R4 and R5 look a bit large to me (I haven't run the numbers)
No they are OK. Re-read the thread to see why. |
|
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 04:12 PM |
|
Joined: May 21, 2003
Posts: 341
Location: Malden MA
|
|
|
Quote:
MBedder wrote:
Quote:
n1ist wrote:
You have two different symbols for non-polarized caps. Unless there's a good reason, I would stick to one
These are just the artefacts of PDF creation.
Look at C3 vs all the other non-pol caps... |
|
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 04:19 PM |
|
Joined: Nov 02, 2009
Posts: 3239
Location: Zelenograd, Russia
|
|
| Ah, I see now. I thought you're referring to C4..C6 vs C1. |
_________________
Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 07:34 PM |
|
Joined: Nov 28, 2004
Posts: 3551
Location: San Diego, Ca
|
|
|
MBedder wrote:
indianajones11 wrote:
You better make sure the FETs even turn on at all because of gate capacitance, B 4 making a board ! If they won't, you'll need FET driver chips .
Why on Earth wouldn't they turn on? Is the 1..2 nF G-S capacity any terrible?
If that's the builtin Cap(gs), it ought to be good with logic level Vins . It's a good "just in case" practice, though . In my case, I was trying to avoid my client expense of the drivers, and the uC 5V just wasn't up to it ( for the FET I was using ) . |
_________________
1) Studio 4.18 build 716 (SP3)
2) WinAvr 20100110
3) PN, all on Doze XP... For Now
A) Avr Dragon ver. 1
B) Avr MKII ISP, 2009 model
C) MKII JTAGICE ver. 1
|
| |
|
|
|
|
|
|
Posted: Dec 20, 2011 - 07:40 PM |
|
Joined: Nov 02, 2009
Posts: 3239
Location: Zelenograd, Russia
|
|
| The OP's MOSFETS are special logic level MOSFETs with a ~2V threshold. |
_________________
Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.
|
| |
|
|
|
|
|
|
Posted: May 22, 2012 - 08:01 AM |
|
Joined: Dec 15, 2011
Posts: 12
|
|
Okay folks. I appreciate the help. I've made a few changes now that I've gotten back to this project and decided to control each of the blinkers independently (2 front, 4 in the back).
Thinking about the MOSFETs having 10k ohms on the gates and 20k ohms on sources is going to end up with a current of 0.0025 watts at the 5 volt output. Is that going to be enough current to control the gates or is it entirely voltage based?
I've attached a PDF of the current schematic.
Also, I've switched the resistors to 1206 for a little more watt handling capacity. |
|
|
| |
|
|
|
|
|
|
Posted: May 22, 2012 - 08:03 AM |
|
Joined: Dec 15, 2011
Posts: 12
|
|
| Oh. Should I pull pins 10 and 12 to ground? Maybe through a 10k? How about AVCC and AGND? Or can I just wire those to VCC and GND? |
|
|
| |
|
|
|
|
|
|
Posted: May 22, 2012 - 12:32 PM |
|
Joined: Nov 02, 2009
Posts: 3239
Location: Zelenograd, Russia
|
|
|
Hardwyre wrote:
Thinking about the MOSFETs having 10k ohms on the gates and 20k ohms on sources is going to end up with a current of 0.0025 watts at the 5 volt output. Is that going to be enough current to control the gates or is it entirely voltage based?
Yes, voltage/gate capacitance based. As has been shown earlier, a 10k series gate resistor slows down the FET switching only by ~20 uS while completely protects the AVR output even if a FET breaks down shorting its gate to drain.
Quote:
I've switched the resistors to 1206 for a little more watt handling capacity.
Not even a funny reason to do this.
Quote:
Should I pull pins 10 and 12 to ground? Maybe through a 10k?
No, just configure them as outputs.
Quote:
How about AVCC and AGND? Or can I just wire those to VCC and GND?
You must do this. |
|
|
| |
|
|
|
|
|
|
Posted: May 23, 2012 - 05:40 AM |
|
Joined: Dec 15, 2011
Posts: 12
|
|
| Awesome. Okay, AVCC and AGND connected. What do you think about opto couplers on the inputs and outputs? |
|
|
| |
|
|
|
|
|
|
Posted: May 23, 2012 - 07:50 AM |
|
Joined: Nov 02, 2009
Posts: 3239
Location: Zelenograd, Russia
|
|
|
Hardwyre wrote:
What do you think about opto couplers on the inputs and outputs?
"God blesses the prudent ones", a nun said and put the condom on a carrot  |
|
|
| |
|
|
|
|
|
|
Posted: Jun 02, 2012 - 01:46 AM |
|
Joined: Dec 15, 2011
Posts: 12
|
|
|
MBedder wrote:
Hardwyre wrote:
What do you think about opto couplers on the inputs and outputs?
"God blesses the prudent ones", a nun said and put the condom on a carrot
I'm going to take that as "killing a spider with flame thrower."
Potentially entertaining, but ultimately overkill. |
|
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
