help needed for 50ms interrupt circuit

[caleb7777]

I was looking at the wrong transistor, The transistor feeding the LEDs fried (T1)

i added a resistor to base and moved led resistor (R6) to before transistor. Plugged it in. Nothing. No power. checked LM7805, fried. hmmm

Current draw fried 5v regulator as well. OK
I have one more replace it and...

it works! for about a minute then shut off. LM7805 fried again. hmmmm

The circuit is drawing too much power. I am sure this can be fixed by adding resistance at some key point but where?
Base of T3??


How can I lower current draw of the circuit?

Thanks

now I need to find more 5v regulators.........

 

[rkosiorek]

just like with W2 i would also add a diode in parallel with W1.

one of the most common reasons for the 7805 to fail is reverse current being forcfed through the device when the voltage at the output is higher than the voltage at the input. that can be prevented by adding a reverse biased diode from the input to the output pin.

i would also add a couple of 0.1uf decoupling capacitors to both the input and the output. if the 12V supply is a dc-dc converter more filtering is necessary 100uf 25V electrolitic on the input.

the 7805 is very well protected for overcurrent and overtemp. i doubt if either of those is the issue.

rick

 

[caleb7777]

rkosiorek I have been tinkering again on this now that I have more components and have been very frustrated by the circuit bernard drew up using the 555. The first beta I built worked great. Still does. But no matter how many times I make the exact same circuit using exact same components I can not get it to work the same as the beta... I know it makes no sense. I gave up and now that I have a breadboard I made the circuit you suggested using a 556 and it does work both ways VERY reliably but as I feared the trigger is always low and the output never stops. How can I modify it so that the triggers on 556 only get a pulse? I tried a passive differentiator using 10kOHM resistor and .1uF cap but it only works on pin 6 if i disconnect the resistor going to the transistor. If I re-add that second half of the 556 the pulse does not generate and there is nothing.

ugh Is there a simple IC that will do it for me?

I was told I need an edge detector to make this work. I was given this schematic but it is Arabic to me.

 

[caleb7777]

I ordered some 4098 IC chips that can be re0wired as edge detectors and would replace the 555 timers entirely. Will try.

 

[rkosiorek]

409 ic? sorry but i am not familiar with that one. could you give me a complete part # or a link to the specs.

i discovered my error with the 555 and 556 schematics. i was using the wrong chip in my breadboard. i used a LM558 which is a QUAD 555. but reading through the spec sheet today i say a major difference in the device. The 558 tigger inputs are indeed EDGE triggered while the 555/556 are level triggered.

what that means is the LM558 input will trigger once and then wait until the switch is opened. the LM555 and 556 will continue triggering as long as the level is held high by the switch.

sorry if i added to the frustration.

rick

 

[rkosiorek]

before you completetly scrap the NE556 circuit entirely maybe you could try a simple mod to make the trigger inputs to the timers AC coupled instead of DC coupled. this should work as an edge-triggered input. ie. only triger when the switch is opened or closed.



it would require the addition of 2 capacitors (C6 and C7) and 2 resistors (R1a and R2a) the capacitors will block the DC signal and only allow through a pulse when the switch is opened or closed

rick

 

[caleb7777]

Sorry I meant 4098 IC

Thanks for the tip on the 558!

That is exactly what I need but I will also try the modded 556 schematic this week.

Thank You!

Here is the 4098 schematic that claims to be an edge triggered timer

 

[caleb7777]

That one does WORK!!

Thanks again rick

I am going to wait and see if the 4098 edge trigger works well before i report a finished design because it would involve very few parts in comparison.
Here is the 556 circuit as It could be used

 

[rkosiorek]

only issue i can see with the CMOS 4098 chip is that it can only source or sink 1or 2mA on the output. nowhere near enough to drive those relays or leds. you will need to use a transistor switch to drive them. that will add a little to the complexity. the circuit would be similar to the circuit around T3 that you are using to drive W1.

looking at your circuit i would add a small diode (1n4148) in parallel to T3. this would be to protect T3 from any kickback from W1 and the 100uF capacitor. it would be connected from the collector to the emitter of T3 with the cathode connected to the emitter and the anode to 0V (reverse biased)

rick

glad to see that it worked. when you are bragging about building it yourself. every tenth time remember to give the forum some credit.

 

[caleb7777]

Never fear even the image has your name "RK"

but if I llike the 4098 better I might brag. cause then I will have designed it all. But I will always owe you one for helping me understand the workings of electronics.

 

[caleb7777]

OK so I have 2 working units!
These are still beta as I have not actually used them on a bike yet but test out as they should.

See post 1:http://endless-sphere.com/forums/viewtopic.php?f=2&t=10209

 

[rkosiorek]

Switches are mechanical objects
Contacts will bounce up and down when moved
Produces a series of short pulses

the 4098 circuit likely is more sensitive to this bounce. and occasionally will trigger twice. the 556 circuit uses those last capacitors that were added on the input to turn them into edge triggered devices. those capacitors also 'de-bounce' the switch and stop multiple triggers.

rick

 

[caleb7777]

Its nothing that fast. The relay activates for a full timing constant and then goes again. But only on the downshift side. (ie CA no longer delivers ground).
But it is fine.

I ordred a few resistors and cas ad will youtube the operation at 1000x so operation can be seen in seconds for entire process.

 

[rkosiorek]

i'll be interested to see it work. i'll be even more interested to see it work in action on the bike.

i'm most interested to see if on the downshift end wheter the freewheeling voltage of the delta winding will be high enough to damage the controller? my hunch is that it will turn out to be much ado about nothing in practice. even though it is valid theoretically.

rick

 

[caleb7777]

Thats the nice thing about controlling by CA. It will only downshift after getting back below set speed. theoretically.