One for the electronics guru's

[BradW]

Also the majority of people who are likely to work on a car know how to use a relay but have never heard of a mosfet so if something goes wrong at least someone can fix it if its kept simple, 99% of cars still use relays to switch indicators, wipers, fans etc for these reasons
cheers

Congratulations, you just described me .
Being an Auto Elec I know all about relays and such but nothing about Mosfet's, pic's or avr's. This circuit is not going into my car so it needs to be something that others can also repair if need be. I also like to keep things as simple as possible.

Going back to my original circuit, I used R3 to give C1 a longer discharge time hence giving a longer off time. I did it this way so that when you turn the intermittent wiper on you get a sweep straight away instead of waiting the delay time for a sweep.
Will it work like this?
Also I read in the relay driving article that a capacitor should be fitted to pin 5 to stop noise spikes in the supply line. How do you work out what size capacitor to use for this?

[camrygt?]

The capacitor from pin 5 should be 0.01 uf or 0.1uf use the little yellow or white MKT ones. The size is not really critical, i usually use 0.1uf and put a 100 uf cap across the power supply near the chip as this helps keep the chip stable when switching.
Your circuit looks like it should work, probably easiest just to build it and have a play with it using a light globe for the motor to start with

[BradW]

Your circuit looks like it should work, probably easiest just to build it and have a play with it using a light globe for the motor to start with

Plan on doing this in the new year. I have to go and spend some money on the gear to do so.
Thanks again for the input, it is greatly appreciated .

[o_man_ra23]

Sorry, ive just found micros to have funny outputs when not told what to do. Mostly applicable to EPLD's and PLC's, but can happen with the interrupt sequences of processors. Also, if you forget to clear a flag, or set one in assembly, then after the first 10,000 chips are programmed, youve already lost your job for not making sure it works. Why do you think car manufacturers make prototypes?? and software manufacturers release beta testing versions?? Its so that things dont fuck up cause of small mistakes etc. Besides that, if anyone is like me, then the things i do for my own car are always having new ideas thrown at them... so updates during design and manufacture are always being made.

My method of doing things anyways... and it works for me... things dont tend to fuck up after they have been extensively tested.

Cheers, Owen

[BradW]

Well, time to rehash an old thread .
I finally got around to making my circuit that I posted at the start of this thread. It was a dismal failure . I think the size of R3 was too big to allow C1 to discharge through pin 7 of the 555 as it would give 1 pulse then do nothing. So it was back to the drawing board. I drew up a new circuit (read, copied the one that ndgcpr posted on the first page and used my own sized components ). I made it up and had some success but not complete success. The timer circuit is working giving me the times I need but the problem I am now having is with the transistor.
I measured the current draw on 4 different brands of relays, all 12v and with a contact rating of 30A. They measured between 150mA to 170mA. After reading the relay driving basics pdf file from the jaycar site I came to the conclusion that I needed a 3K3 ohm resistor to give 3.4mA base current to the transistor to drive the relay. I ran the circuit and measured the output voltage from the transistor with a relay connected. When the output of the 555 is high I get 3.5v out of the transistor and when the output of the 555 is low I get 5v out of the transistor. This is the problem as it is not enough to drive the relay correctly. Can anyone give me any idea as to what I am doing wrong here. I have attached a diagram of my circuit. I am running it off a 13.8v power supply for testing.

[camrygt?]

If you look at the post I made before and the link to the specs on a normal (not cmos) 555 It has a pair of transistors built into it which can safely sink or source 200ma so just connect it straight to your relay. But use diodes to stop back emf

[BradW]

That would be good but the problem I have is the output is high for approx 9 secs and low for approx 2 secs which is the reverse of what I need. I am thinking about ditching the transistor and using the NC contacts of a small relay mounted on the board to trigger the low speed wiper relay on the car. The other thing I'm not happy with is there is about 12 secs wait after it is turned on before you get the first sweep.

[Toobs]

What about one of these bad boys:
http://www.jaycar.com.au/productView.asp?ID=KC5379

This is a very sophisticated timer which can be adapted for two different types of uses. The first is ‘one shot’ operation, which can be used to keep electric windows active, or a thermo fan running for a period after ignition is switched off etc. The second is a ‘pulse’ type operation, which can be used to squirt water spray for 1 second every 9 seconds. The uses are endless, and the time is adjustable from 0.1 seconds, to 16.5 minutes via easy to use (and accurate) digital switches. Kit supplied with PCB, and all electronic components.

[BradW]

What about one of these bad boys:
http://www.jaycar.com.au/productView.asp?ID=KC5379

That would be perfect, but where is the fun in that when I can spend endless nights banging my head against the wall trying work out how to get my circuit working .

Thanks for the link Toobs, I may end up buying one of them instead.

[camrygt?]

Ah thats easy fixed. The output of the 555 has a pair of complimentary transistors in it. So when it is off it will pull to ground...so join your relay between 12volt and the output of the timer. then it will invert the logic of the relay

[BradW]

That's a very good idea. +rep for you .
Although I would still like to know where I have gone wrong with this transistor setup, if only for future reference.

[seven_mgte]

didnt read the whole thread, or really any of it but this is what a guy did for an aw11. might be relevant. sorry, it dosn't have pics.

Well I've done it. Fully variable intermittent wipers for a MK1 MR2. Today was a great opportunity to test it too in Vancouver's drizzly rain. If you can solder and have the patience of a monk, you too can have this indespensable feature on your 2.

If you use the same components that I used, you'll be able to vary the frequency of your wipers from half a second (faster than current speed), to "Is this still on?" I timed it to 5 minutes, and still it didn't sweep. So it's likely somewhere over 5 minutes but (hopefully) less than 10. I'll edit this message when I know how long.


Here's what you need...

- 1Mohm, linear taper, 1/2W potentiometer (AKA: pot - don't get excited). $2.00

- 24Kohm, 1/4W, 5% resistor. $0.25 (but might have to buy more than one)

- knob for the pot, suitable to your tastes. $2.00 depending on taste

- 8" of 24-22GA stranded wire. (cheap)

- These instructions and accompanying pictures. (FREE!)


How does it work?

Inside the wiper switch assembly, there is a small printed circuit board. This board is exclusively for the intermittent setting of the wiper switch. The timing circuit is a standard R/C network and a couple of transistors to drive the relay.

To change the speed from the preset to variable, simply replace R1 (91Kohm, 1/8W, 5%) with a pot. The extra 24Kohm resistor is so that when the pot is adjusted to 0 ohms, the circuit is not completely shorted out, you can vary this resistor from about 20Kohm to 100Kohm. This resistor will determine the fastest speed setting.

I used a 1Mohm pot, but the slowest setting is well over 5 minutes (still don't know what the slowest is exactly). If you feel that's too long, try a 500Kohm pot, but I don't know what the slowest setting would be. The type of pot you use will determine where you place it. Mine was part # 62-229-1 from RP Electronics in Burnaby. It was the smallest one available, because I wanted to fit mine inside the existing switch assembly. It also came with a switch, which I cut out (see pics G & H).

If you decide to put your pot somewhere else, you'll need more wire, and maybe a connector so that you can disconnect it in the future for servicing other parts of your dashboard.

Polarity is irrelevant here so don't worry about it. The only thing to be careful about with the wiring, is direction of the pot from fastest to slowest. Use an ohmmeter to test which way you want it to go. I set mine so that the fastest speed (lowest resistance) is full counter-clockwise (see pic J). If you want the opposite direction, just reverse your connections at the pot.


Here's how to do it...

- Read through these instructions and look at the pictures first as I included techie-tips on my experiences.

- Disassemble the dashboard area that houses the wiper switch.

- Remove the wiper switch from the housing (see pic A).

- Remove the one screw at the top of the switch (see pic B).

- Pry up on the sides, in the middle of the top cover of the plastic case on the switch, until the two tabs at the front of the switch let go.

- Gently pry up the circuit board the same way as the plastic cover.

- Locate R1 (91Kohm, 1/8W, 5%) in the middle of the board (see pic c & D). Unsolder R1 and discard it (see pic E).

- Cut your 8" of wire in half.

- Solder both wires to the pot (see pic J)

- Solder the free pin of the pot to the center pin. If you know electronics, you'll know why. If not, just do it anyhow.

- Mount the pot to the switch housing (some measuring involved here, you'll drill a 9/32" hole in the side of the switch, make sure of the position, you only have one shot at doing it right (see pic K).

- Just finger tighten the nut on the pot for now. Run the wires under the shaft of the existing switch. (see pic K).

- Solder one of the wires to one of the holes of R1. (see pic L)

- Solder the other wire to one end of the 24Kohm resistor.

- Solder the 24Kohm resistor to the other hole of R1.

- Electrically, you're done.

- Test fit the board to make sure it doesn't pinch the wires or the resistor as you close the case (see pic M). Leave the top of the switch open for now.

- Prepare the housing for the switch. Measure the area to be drilled for another 9/32" hole. Again like the first hole, make sure of the position, you only have one shot at doing it right (see pic N). I realize this is repeated, but measuring out before drilling is extremely important!

- Mount the modified switch into the dash housing. This part is easier said than done! A lot of expletives and one wasted pot (good thing I bought two), before I had it installed in the housing (see pic N). TIP...remember when I said to finger tighten the nut for the pot? Now loosen it completely, but keep the nut/washer on the shaft of the pot. Also remember when I said to leave the top of the switch open? This will help when you retighten the nut on the pot.

- Line up the assembly into the housing, the loose pot and your accurate measurements will help. Once in place, put the nut on the front of the switch, but leave it loose. This will help when you tighten the pot down, between the switch housing and the dash housing. Check the inside of the switch to make sure that as you're tightening the pot, it isn't rotating on the inside. Once it's tightened, then check the wires and pot, to make sure that they aren't "stressed". Close the case of the switch (have fun with the screw). Tighten the front of the switch.

- Re-assemble the dashboard (see pics P & Q)


NOTES:

I started this project late yesterday, but needed my car to get parts this morning. So I had to re-assemble everything (partially, see pic F). Make sure you have all the parts in advance, and give yourself about a day to complete this task. You have the benefit of knowing which resistor to pull out, something that took me about two hours to figure out. I wanted to make my pot fit inside the switch housing, to make it look cleaner on the dashboard. If you want to avoid some headaches, consider placing the pot somewhere else on the dash.

Pete

[camrygt?]

That's a very good idea. +rep for you .
Although I would still like to know where I have gone wrong with this transistor setup, if only for future reference.

the relay driving basics thing on the jaycar website (under primers heading) has a bit of imformation on seting up transistors, but the 3 or so K resistor you mentioned should work Ok. Either PNP or NPN transistor will work just the logic will be inverted.
for a typical NPN transistor BC337 or similar the pinouts are collector, base, emitter looking from the flat side. Connect the emitter to ground. collector to relay coil/other side of coil to power. then the base through the 3 or so k resistor to 555.
Don't forget a back emf diode across the relay coil

edit just looked at your diagram swap the collector and emittor around
Also a capacitor connected between the base of the transitor in your circuit and ground (after the 3k resistor) might give you a pulse when you first turn it on. try some values around 2uf - 500uf

[BradW]

edit just looked at your diagram swap the collector and emittor around
Also a capacitor connected between the base of the transitor in your circuit and ground (after the 3k resistor) might give you a pulse when you first turn it on. try some values around 2uf - 500uf

So the emittor should be connected to 12v, the collector to the relay coil and other side of relay coil to earth?
The cap would go + between the resistor and base of transistor and - to earth, correct?

I really appreciate the help and apoligize if these are dumb questions.
More +rep coming your way after I "spread the love" .

seven mgte: thanks for the input but the car I am fitting this to doesn't have intermittent wipers at all.

[camrygt?]

yep that all sounds right.

The capacitor is just an idea it may not work. My theory there is that it will be discharged when the circuit first powers up and pull the base low turning the PNP transitor on for a moment until it charges up.
It will also cause a bit of a delay when the 555 changes state so if it does appear to work keep the value of this cap lower rather than higher