SX- and Mosfet drivers

datacps · 2007-07-07 03:02

I read in the fuel injector thread that the IRL510 had a logic level gate drive. Does that mean I can drive that·mosfet directly from my sx pin?
Do I put a resistor in series?
Should I·use a mosfet driver?
I saw in Markystuff circuit that he was driving the mosfets directly off the SX- Chip..
I am planning on driving a switching power supply with 12v
The output is 3KV· 10ma to run neons ect..
I am worried if I don't use a mosfet driver to isolate the chip I may get a kickback high enough to damage the chip.

·I am also going to power my sx from the power supply through a·RJ11
and about 4 ft in length. That way I can plug in the driver·unit and power my controller·from one power source so for this circuit I may need the mosfet driver or a·OP AMP

Also in my final design·should I ·place a 10K R on the mclr pin 28·to·vdd·and use the caps across vss and vdd close to the chip and the programing header for the SX Key?
I have my schematic in Express·and I am waiting for the encoders so I am not finished with it yet. The driver for the transformer pretty much is finished and if I can eliminate the mosfet driver·for my other·project that would be great.
I am using a 12v regulator to feed power to the control unit through the 4 ft cable and then I have the 5 volt regulator powering the·SX chip. I am getting ready· to make the boards as soon as I get the encoders and test the program and I don't want to make a mistake on the first run.. Any advice would be greatly appreciated..Thanks

datacps · 2007-07-07 14:33

I am posting a bitmap . This circut is simple and there is so much information on this site. I was searching for certain rules to follow when developing projects with the SX and I noticed different opinions about where to place caps and resistors to eliminate glicthes. Thanks again...

datacps · 2007-07-07 17:43

I don't think the values will show on the other screen shot of my schematic. So I am displaying both circuit close up to be able to read the values..

LoopyByteloose · 2007-07-08 08:25

From what I have gleaned of general principles, one can easily drive a MOSfet directly from an SX pin if designed for 5 volt operation.

Also, there seems to be a lot of mythology about damage.

The primary reason for the resistor between the two is to protect the MOSfet from transient damage to its input pin. This might occur if the SX was unplugged and someone was to touch the input while having a static charge. As you can see, the damaging event is a bit far fetched. Most MOSfets are fine once they are in their appropriate circuit. But they appear to be easily damaged by breadboarding and hot plugging.

Adding a resistor at certain frequencies may in some cases create a timing delay in conjuction with the MOSfets internal capacitances. But if that isn't an issue, the resistor doesn't hurt the circuit, merely adds to the component count.

At the end of the day, if you have a MOSfet rated for digital I/O control, you can forget it. But static electricity still will remain an issue during assembly.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"Everything in the world is purchased by labour; and our passions are the only causes of labor." -- David·Hume (1711-76)········

···················· Tropically,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

Post Edited (Kramer) : 7/8/2007 8:32:09 AM GMT

datacps · 2007-07-08 14:51

Thanks Kramer
···· That pretty much answer that. I guess maybe the only thing I need to worry about is the 4 ft of cales that will be between the driver circuit and the·control circuit.
·· On your advise do I need to worry about voltage drop affecting the mosfet and maybe put an op amp in the drive circuit·to keep the precise control voltage at the gate pin? Or in this case does the mosfet driver or op amp need to be there. The freqout pin to the·RJ11 is 5·on the schematics?
What do you think?
Thanks a ain

·

····

·

Mike Green · 2007-07-08 15:10

4 ft of cable is a lot for an ordinary logic signal. The problem is not so much voltage drop since the MOSFET gate is high impedance. The problem is the capacitance and inductance of the cable and the problems that will cause with rise time, ringing, overshoot, etc. In addition, you will have all the electrical noise in the vehicle running near the 4 ft. cable and inducing noise spikes in the control line, possibly feeding back into the SX circuit and falsely triggering the MOSFET. You'd be much better off keeping the MOSFETs close to the controller and running the high level injector current through adequate wires to the injectors.

You could use lower level drivers, like those for RS-485 or RS-232 between the controller and the drivers. These slow down the control signals so they're less vulnerable to noise among other things. You could do the same sort of thing with op amps and a few passive components.

datacps · 2007-07-08 15:55

Good I dea Mike I am going to put the mosfet on the driver side and just run the 12v pulses out to the transformers or the coils..

So I will use the IRL510 close to the chip and run the output to the driver circuit.
I don't know why I just did not think of doing it that way fron the start..
Thanks again that should resolve my issue and is the answer I was looking for..
What about placing the caps and resistor across the vdd and vss close to the chip and on the SX- Key programin header? how should i put them on the circuit..
Thanks Mike

Capt. Quirk · 2007-07-08 21:20

This thread is very timely for me, I was thinking the same thing (the SX about 4 feet back and the IRL510 right·next to·the Parallax, Melexis Digital Hall Sensor for my high energy TCI style ignition) In order for a TCI ignition to create a high voltage spark, it needs the coil supply voltage shut off very fast. The IRL 510 is fast at 15 ns according to the data sheet, But this old thread about sparks and the SX have also been a concern

The normal setup in a car would be with the sensor on the crankshaft and the controller about 4' to 6' away. But this is in one of my airplanes with an 80cc twin. The coil should put out better than 35 Kv and is easily twice the normal voltage output of the CDI systems used for RC Airplanes, so I was considering building a "Farraday Cage" for both the SX and the RC receiver with the antennae going out the back of the airplane.

Does anybody have any other ideas on how I can make this system work, in a truly delicate envionment and help keep from having a $5000.00 model airplane go SPLAT! (it has the latest 14 ch Futaba radio inside)

I am finishing up the ignition test stand today and I should be testing the ignition tonight or tomorrow.

Thanks

Bill

datacps · 2007-07-09 00:59

Thanks Quirk I read the entire thread. The AN-3 was very helpful. I am going to put a GRN plane under my chip and run pin 4 to it like it states. I also got to put a 10K resistor from MCLR to VDD.
·· I am glad that I posted.· I am also going to include the AN-3 for anyone else that may need the info so they don't have to search on the other threads to find it.
I don't have a solution for your concern but you did help me with mine..
1. Current and voltage peaks are the most common causes
of EMI. Use of proper decoupling capacitors between
Power and Ground at the power source to the
board as well as on the SX Vdd and Vss pins may help.
It is also important to evaluate the proper capacitor
type, frequency response, placement, output load effect,
and size.

Thanks goes to you Quirk and Mike Green

SX- and Mosfet drivers

Comments