Pulse Stretching

Hi folks. - old thread, sorry. New to the forum and electronics and I have an issue I need help with - on a Kawasaki motorbike. The author of the thread below did a brake upgrade with I am also doing. Everything worked except the LED warning light. His final post (four and a half years ago) gives the solution, but it's meaningless to me as I am a mechanic who's not into electronics - yet. He says a pulse stretching circuit ( known as a gated monostable) was the answer. I can see the LED flash when the ignition is turned on but it needs to stay on until the ECU extinguishes it. Any help greatly appreciated.


http://kawasakiworld.com/zx-14/43096-zzr1400-abs-brake-wiring.html

 

Hi Freddy,

Welcome to the forums

I suggest that you start your own thread and provide a clear description of what you have to work with and what you are trying to achieve. Whilst sometimes old threads are relative to your situation, and can give you some good information, technology does move on and so do members and ways of doing things. Your best chance at a successful outcome here is to take advantage of the "current" way of thinking. There's nothing wrong with posting a wee link to an old thread, in your own new one, that has some information you think is relevant to what your trying to do.

 

Thanks Mr Wolf. Should I do so in this Electronic Projects section or in the Automotive Electronics area?

 

No need... I have moved them here!!

 

Can you give a link to the manual and to the circuit you are trying to follow? (I did not find the manual on the link you gave.)

John

 

Thanks John. Although I have the service manual for the bike it doesn't give much away in the way to circuitry. The author of the topic in my first post discovered the info in the par starting 'Basically.......', but I don't know where he got it from.

Hi
I thought I had better close this out in case the information is ever needed by anyone else.

The answer is that the blk/wht wire to the meter unit is ACTIVE LOW.

Basically the indicator LED for the ABS system in the meter unit is connected to the positive rail of the battery (12v). In order to make it illuminate and signal the rider, the ABS ECU momentarily drops the blk/wht wire to almost ground (almost 0v ) which lights the indicator and passes information about the state of the brake system to the rider.
After a bit if digging in the Kawasaki ZZR1400 manual this confirmed in the electrical section 16.69 where pin 6 of the meter circuit is listed as ABS LED ( - ).
Lastly, the active pulse to illuminate the ABS LED is of a very short duration. So short as to be impractical as an indicator.( around 0.01 seconds). This was solved using a pulse stretching circuit ( known as a gated monostable) where a short signal in produces a longer signal out. In the case 0.01 seconds in makes 0.5 seconds out.

Curiously the short pulse only occurred at start up. The diagnostic message pulses were well timed without any pulse stretching. These pulses are shown as 0.4secs in the manual.

The outcome of this is that I now have a ZX12R with ABS. Fully tested and working and now with the added (useful) feedback of the LED fault indicator.

Have to say that the ABS setup you ZZR guys have enjoyed as standard is very good. Never had ABS before, but can't see me doing without it again.

The last sentence in my first post is where I'm at, which was where he was when he asked his question which he eventually resolved. I can add the page he quotes but it's no help other than to show pin 6 is ABS LED (-) in the back of the instrument cluster.

 

Using a short pulse to initiate a longer pulse is common. The combination of a short negative pulse to trigger the longer pulse suggests the circuit may be based on the venerable NE555 /LM555 chip, such as here: https://electrosome.com/monostable-multivibrator-555-timer/

It is not necessary, however, to use that chip. A search for "monostable" will show several circuits, including ones made from discrete devices (i.e., transistors, capacitors, resistors). The 555 is a durable chip and can operate 12V well. Plus, there will be lots of help available for it. So, while it may lack novelty, it may well be your best bet. You may need to condition your result to stop unwanted blips of the LED when power is first applied, but before the ready signal is received, if such blips will be a problem. None of those blips will be on the order of 0.5 to 1 second, so they may not be a problem.

John

 

Thanks very much John. Regrettably, it's just about 'all Greek to me' as someone famous once said. Leaving aside for a moment the components that are in the circuit diagram, allow me to add further info on what the LED is supposed to do, which you no doubt know anyway. It's supposed to illuminate when the key is turned on and stay on until a couple of seconds after the bike moves off - providing there are no faults eg, flat tyre, and stay off until a fault is detected. Will the venerable system you describe behave in this way?

Secondly what 'bits' do I need to buy to make it up, or can 'it' be purchased as a unit?

Thank you for your help and tolerance of an electronics dullard.

 

Here's a simple discrete one-shot circuit (as an alternate to a 555 circuit).
The values shown give a pulse of about 1 second, which can be varied by selection of the value for C1.
Rtest is just for simulation purposes and should not be needed in the actual circuit.
The transistors and diode can be any common small types.
The resistor values are not critical and can be within 10% of those shown.
The output will stay low as long as the input is low and then will go high after the delay when the input goes high.

 

Thank you crutschow. My understanding of electronic circuit diagrams to low but I am familiar with automotive 12v circuits so please bear with me.

Will that circuit do as I stated in post 8: Leaving aside for a moment the components that are in the circuit diagram, allow me to add further info on what the LED is supposed to do, which you no doubt know anyway. It's supposed to illuminate when the key is turned on and stay on until a couple of seconds after the bike moves off (after the ABS ECU has detected no faults) - providing there are no faults eg, flat tyre, and stay off until a fault is detected. ?

I also do not know what signals the LED to extinguish when the bike moves off after the system detects no faults. Perhaps you or someone else may. If so, how? Will your circuit enable this, or the 555, or.........?

Secondly, the original author, who's work I'm trying to emulate, sates in post 6: Basically the indicator LED for the ABS system in the meter unit is connected to the positive rail of the battery (12v). In order to make it illuminate and signal the rider, the ABS ECU momentarily drops the blk/wht wire (which runs from the meter/cluster LED to the ABS ECU) to almost ground (almost 0v ) which lights the indicator LED and passes information about the state of the brake system to the rider.
After a bit if digging in the Kawasaki ZZR1400 manual this confirmed in the electrical section 16.69 where pin 6 of the meter circuit is listed as ABS LED ( - ).
Lastly, the active pulse to illuminate the ABS LED is of a very short duration. So short as to be impractical as an indicator.( around 0.01 seconds). This was solved using a pulse stretching circuit ( known as a gated monostable) where a short signal in produces a longer signal out. In the case 0.01 seconds in makes 0.5 seconds out.

The blk/wht wire at pin 6 (rear of meter/cluster) is momentarily earthed by the ABS ECU. Is that what your diagram is showing at the left side?

Thanks in advance to helpers.

 

The left input is the momentary low signal from the ECU. That is what the Vecu source (round circle) on the left is simulating.
The right output goes to the meter cluster to light the LED.
Obviously you need to break the present connection between those two to add this circuit in between.
(This is for how I understand it works).

 

Thanks crutschow. Ill leave this issue aside of the time being while I concentrate on the hydraulic and mechanical installation. From bench testing, I know the unit is active and switches on with ignition, just the LED I'm trying to retro fit does work as hoped. I'll be back............

 

I found the problem that's folks. While trying various combinations of connections for the LED while bench testing, I stumbled upon the light working.

The ABS unit has power supplied via 3 fuses. One for the ABS pump, one for the ABS valves and one for the ABS ECU. What I had been doing was connecting the first 2 to the battery and touching the last one on the +ve terminal to imitate turning on the key. Doing so, the LED never came on tho it did flash. It trying various combinations of connections I found by accident that if all 3 leads were connected to the +ve terminal then the earth lead from unit was touched to -ve, the LED came on and stayed on. Don't know why, but who cares? I don't.

You beauty!!!!

I only had one sensor plug into the harness, as the other was fitted to the bike yesterday, and when I put the ABS into self-diagnosis mode the LED flashed code 44 (missing sensor). SUCCESS!!!

Thanks for the input folks. And I really must study up on electronics when I get some more time.

 

Well not so fast buster! I was a premature in saying that I had resolved my problem.

The LED does work as it should but only when the ABS ECU is put into self-diagnostic mode. The ECU causes the LED to flash in sequences that can then be compared to the chart in the w/shop manual to identify the fault.

Some I'm back to square one . The the ignition is turned on in normal mode the LED gives a single flash as described by the person (from post #10 above) whose work I am trying to copy: Lastly, the active pulse to illuminate the ABS LED is of a very short duration. So short as to be impractical as an indicator.( around 0.01 seconds). This was solved using a pulse stretching circuit ( known as a gated monostable) where a short signal in produces a longer signal out. In the case 0.01 seconds in makes 0.5 seconds out.

So John or crutschow, precisely what components do I need to solve this please? And what is the easiest way? Thanks

 

You'll have do decide what's the easiest way, as determined by the required parts for each approach.

For precisely what parts in my schematic do you have a question?

 

Thanks for the reply cruts. My dear old dad used to say 'you can't argue with ignorance (or is it ignorants).'

I went to the local electronics shop yesterday and they sold me $4.75 worth of bits to make up a timer using the venerable 555 thingie. I'll have a play with them over the next week to see what happens and post my results.

 

I had a good look at the diagram which is in the link in post 7 and have two questions. Firstly some info: the trigger pulse starts lo, goes hi then returns to lo. This is the opposite of what's shown at the left of the 555 timer diagram. Will it work if wired as shown?

Secondly, the 2 capacitors I was given are very different in size. Is the larger one a C or C' as shown in the diagram?

Thanks folks.

 

Your original description said "Basically the indicator LED for the ABS system in the meter unit is connected to the positive rail of the battery (12v). In order to make it illuminate and signal the rider, the ABS ECU momentarily drops the blk/wht wire to almost ground (almost 0v ) which lights the indicator and passes information about the state of the brake system to the rider."
This is the opposite of what you are saying now, so which is it?

Yes, the 555 circuit requires a trigger pulse that starts high and goes momentarily low.

Can't tell which cap is which without knowing what's written on them.

 

The problem is my terminology I'm sure. Sorry!

What he said is correct.

 

Given that I have 3 wires on the bike to connect (12v power supply from ignition; earth; and pulse wire from/to ABS ECU) plus the 2 LED wires, which terminals on the board below should be connected to which wires? With thanks.

http://www.ebay.com.au/itm/131044321757?_trksid=p2060353.m2749.l2649&ssPageName=STRK:MEBIDX:IT