UkLegacyT

thanks for that john, id looked on that site when you mentioned the dawes device bleed valve, but hadnt spotted the afr meter. looks like ill buy one of them, as they are slightly cheaper than the luminition ones i have seen for £110 !

thanks again

ian

jmca

John,

All three q's answered! Much appreciated!

Johny.

jmca

Don't want to keep asking mayeb what seems daft questions but there was a mention of simulating the lambda sensor output with a 22K pot into pin 5.

Could you use a 4K7 pot to reduce the voltage of a 1.5V finger battery to simulate the range of readings from 700Mv to 1V of the lamda sensor & connect the output of the 4K7 pot to pin 5?

Regards

Johny

john banks

Yes. It is just a way of making a test voltage that is all. Any suitably large enough pot that will not flow too much current is fine. Use I=V/R, and say go for 1mA or less say.

Hanslow

Should not be beyond someone.....go on then John On second thoughts, I will only get confused again

Someone that understands electronics maybe

john banks

I'm still trying to work on the FCD and the EBC has not seen any action for the last month or so. I know, just give up the day job

Hanslow

John / Scott,

Quick question about Scott's layout diagram,

Which LEDs would this use (from the chip side of things)?
Just been looking at the diagram again, and given my little electronics knowledge (i.e. GCSE/A-Level Physics that has all been forgotten ), is it using the pins for LEDs 1,2,4,6,8 (assuming 10 LEDs)?

If so, I'll probably need to change my reference voltage, or redo it, cos I wanted pins for LEDs 6,7,8,9,10 (I think). Basically I wanted to monitor 810mV to 890mV inclusive, i.e. 5 LEDs in 20mV increments.
Assuming it is using 1,2,4,6,8 I'm assuming it will be referencing 710, 730, 770, 810, 850 (if pin 4 is 690, pin 6 is 890)?

Is this right? Or do I need to redo bits? Cos I want 810, 830, 850, 870, 890mV. I've either just confused myself, or slightly misunderstood the diagram


[Edited by Hanslow - 3/2/2002 7:46:12 PM]

Hanslow

Yep that's the sort of thing I was after. I think looking at Scott's diagram, it doesn't use a contiguous block of pins does it?

I think it jumps a coule here and there? If so, I guess I just need to change which pins it is running from by covering up my track cuts, and cutting some different ones?

john banks

See I have edited above - does it make sense?

john banks

You have to imagine all ten are connected to work out the reference points. Note that a lower reference of 800mV does not mean that the 1st LED lights here - it is 10% of the spread higher - since our spread is 800-900 mV if we connected all 10 we would have 810 820 830 840 850 860 870 880 890 900 for LEDs 1-10 respectively. Since you have 1,3,5,7,9 connected you get 810 830 850 870 890. As Spok or whoever he is might say, simple logic. However, my brain now hurts

Hanslow

Your brain hurts

You don't want to know what mine is like then

After your post explaining the pin numbering (that I had stupidly counted the wrong way round when double checking) and explaining the reference voltages, it all made sense

I'll look into getting a 220k pot to test the circuit fully when I've wired the LEDs in. Looks to be OK now you've explained that I can't count pins correctly

Thanks again for the assistance, it was very helpful

Hanslow

John,

Just been a complete spoon

Was testing against a different voltage (12v instead of 9v) to see if there was a difference in values. Had the current set too high

Thought I heard a 'click' somewhere....now it's dead...that'll teach me to fiddle.

Is it likely to be one of the POTs or the chip? Is there any way I can test what has knackered? I'm no longer getting any readings on pin 4 or 6

Nothing like publicly humiliating myself

john banks

It will be the chip. Thankfully with this circuit that is about all you can blow up. Don't worry - it is irritating and you end up spending more time, but we all blow things up - should see the FCD thread and me actually smoking it TWICE and then giving up. Good for the soul probably. Thankfully this one is only £4.

Now, I socketed the chip as I thought I would blow it up. Whenever I socket a chip I don't blow it up, when I don't I do. I can see you haven't and it has so that proves it

The 220K is not crucial - it could be anything over 5K I reckon.

The other question I want to know is WHY it blew up? 12 V supply should be fine - if you see the data sheet it goes MUCH higher.

Hanslow

Dunno, my 12v supply was reading 1.5 amps, whereas the 9v was reading about 0.5 amps. It was just one of those things where I thought it will be 12v in the car so I'll give it a quick go, famous last words

I was heading down maplin to get the pot for testing anyway Looks like I'll be getting a bit more.

Might invest in a socket as well then (for extra safety ) and rewire it all.

On a personal note, do you actually ever go to sleep?

john banks

I get a good 8 hours a night - have to otherwise I am useless.

Current should go up roughly in proportion to voltage - so under 0.7A. This still sounds very high - I don't expect it would draw over 100mA with all LEDs lit? Perhaps you adjusted one of the pots too far and the LEDs went very bright and overloaded the chip?

Hanslow

I haven't wired the LEDs up yet. I've only been testing the outputs on pins 4 and 6 against ground so far. Didn't know if that mattered? What sort of current does it pull in the car?

Will the values still be the same on pins 4 and 6 if set up against a 9v supply and then moved onto a 12v supply, or will there be an extra bit of tweaking involved once it's inside the car? If they are the same then I am going to wish I asked this question before

So are you saying that 1.5amps is too high and is likely to have blown it?

john banks

They are roughly the same, but I tweaked them slightly in the car. I tested mine with a PP3 battery, from conception through two designs and only flattened one battery, but I did not measure the current drain.

Is you PSU actually showing how much current is passing, or is that just the setting you put it on as a maximum?

I would double check your board for shorts. I couldn't see any on the photo you posted - with the chip out, continuity test between adjacent connections to make sure they only flow if they should and a bit of solder is not causing the mischief.

Hanslow

I guess I'll have to do the car tweaking bit then once its in

I just my multimeter across the terminals on the PSU to find out what it was running. Only did this after it popped though 9v was showing 0.5amps and 12v 1.5. I'm assuming there was just too much current for the chip and it fried. Be interested to find out what it draws in the car.

Well the board was fine last night on the 9v battery. Got up this morning, though I would try 12v and pop, dead board. Don't think there are any shorts. I checked the board over after it died and nothing seemed to be connected to what it shouldn't be.

Not to worry, I'll scoot down Maplin and just get some more stuff. I'm going to wire up a completely new board, cos it's a pain trying to get the chip out with all them legs soldered. I'll get a socket this time as well, then if I pop it again, I can just ping it out. Got plans for today, but hopefully might be able to have a go tonight. I'll let you know how it all goes

Thanks again

Steve

Hanslow

Quicky John (or anyone else )

I've wired it up now with my new bits, and done the LEDs as well. I have another 9v battery, a 220k pot and lots of wire

How do I wire up a test circuit on some breadboard (can't be arsed to solder) so that I can get a test signal onto pin 5 on my main board?

I know what potentiometers are used for but I haven't a clue how they actually work.

Sorry for being a pain and a bit of a thicky

john banks

Pots have three legs. Two are at the opposite ends of usually a carbon track which is a resistor. Usually the middle one is a wiper that rotates along the track. So connect one end to 0V, the other to 9V. The wiper will get a voltage proportional to the position on the track of the 9V if it is a linear pot, and proportional to the log of the position if it is a log pot.



On this picture the middle one is in the foreground and is the point which you run to pin 5.

Hanslow

Hmmm, that's what I tried but didn't get anything to happen on the LEDs As far as I could tell, there didn't appear to be any signal voltage actually coming off the middle pin......

I'll go and get my stuff up here with my PC so I can have a fiddle and ask questions

john banks

Best give me a ring. I'll email you my number.

Hanslow

You are a star (again )

Having fed the test circuit off the same battery (i.e. not the dead one ) it all worked

Barely saw the lights cos the 220k pot made the input signal a very fine range. Once I saw them flicker though I twisted it back and they lit up in sequence. The missus wasn't very impressed though

Thanks again for the support, a beer (or a coke) will be on it's way at a mutual meet

Eeee, I'm dead chuffed I'll not blow this one up

I can go to bed now

john banks

Wives never are impressed with this sort of thing, and a change from 13 to 19 PSI was greeted with, "I suppose it is a bit faster now, but it is a bit scary as it just takes off whenever you press the accelerator without any delay."

I think she meant there was less turbo lag

Scott.T

iTrader: (2)

Well done Hanslow, glad to see someone using my stripboard design.

Dead easy to put together, unlike my origional which used all 10 LED's with all LED's and POT's hardwired to the Chip.

This design only takes 10-15mins to build. It takes longer to find the parts in the catalogue.

P.S It's easy and neater if you use tin/copper wire for the linking.

Enjoy.....

Hanslow

LOL @ John I know exactly what you mean.

Scott, your design was a great help. If it hadn't been for my stupidity in understanding the pin numbering (which thankfully John pointed out) I would have attempted a slight re-design, probably with drastic results Thankfully I didn't and followed it exactly and it all worked.

It was great for me as it was like guitar tab is to real music, i.e. readable and usable . At the moment I don't need to know what it does or how it does it, just where to solder what bits So thanks for putting into simple terms I now want to start understanding a little electronics so that my head won't hurt as much on the next DIY job

john banks

Well if you are up for it Fuel cut lifter and electronic boost controller are well underway

Fuel cut lifter should only be a little more complex than this one but two chips to blow up

Ken Ross

Can someone confirm the colour of the wire on an 95 STi, I thought it was White with silver on it, but there is two ! One 5th up on the topof the connector, one 4th up on the bottom of the ecu connector.

I think I may have got the wrong one, as it only reads 790mv on full boost and can read -100mv on over run ! Or is my Lamda sensor on the way out ?

john banks

Could be on the way out unless you are picking up a 100mV offset in the ground wire? I use the ground for the sensor.

I hear you can get lambda sensors from Halfords for £50 might be worth looking at if it is duff, assuming the wires are the same.

Do you get the "Knight Rider" effect of voltage flicking from 0 to 0.7V at idle after warming up?

Ken Ross

I took the ground from the cig lighter, but also tried it earthed to the ignition key surround, will check the offset later, any idea what connector it is on a 95 for the sensor ground ?

As for after warm up, yes it flicks around from around 50mv to 700. Also sits at 280mv during warmup.



[Edited by Ken Ross - 3/4/2002 7:52:30 PM]