somberg

Good info Pat, thanks.

a link to a sensor that bolts under a spark plug:

http://www.pcb.com/products/custom/cus140m05.html

__________________
Bananaman..

Hugh Keir

Pat,

Thanks again for the extra information, Optrand was the manufacturer of the in plug sensor, but as of my last conversations with them about a year ago, a controller was not availble from them to integrate with the sensor to control the ignition timing.

Your comment To stave off detonation AND make better power, you need a "real" octane booster such as TEL, Benzene, Toluene etc

Needs to be expanded to say that the energy in these additives is no greater than 98 RON fuel. Only by altering engine parameters like ignition timing and higher boost in conjunction with the higher octane fuel can you release more power.

To find a fuel that by itself increases power, you have to look at the race fuels as provided by ELF which most would consider prohibitively expensive.

Peter,

Am working away from home right now and am not sure which maps are available from the hand controller.

Somberg,

This is the one I had seen previously and is more real world as the voltage can be read directly. The downside is that you have to recalibrate each time you tighten your plugs, you will get a different pre-load and a different electrical start point for the voltage output. Combine this with the output from the ignition trigger disk and you can do closed loop ignition per cylinder to give maximum cylinder pressure at 16° to 18° after TDC. However it's not as far as I am aware been done yet by any of the Motec's or Pectel's, it's still in the evaluation stage by the big car manufacturers.

There is also Ion sensing, which I believe is standard on SAAB cars and is able to detect knock by looking at the resistance across the plug gap. Here is a link to http://www.meaa-mea.com/products/p_ionic.asp a system that Mitsubishi are working on to do the same thing. I think it might be ion sensing that Autronic are working on.

pat

Peter,

I would certainly suggest that, in the absence of being able to alter the ignition timing curves in an engine management system designed to run on a higher quality fuel, such as Japanese import vehicles, the addition of the "magic bullet" octane boosters will allow the engines to continue to run the advance profile they started with at the factory without the same level of danger of engine demise as running without. It is also not unknown for a knock sensor to go faulty and for this to lead the ECU into a "false sense of security"; given the combination of poorer fuel and an ECU which cannot "hear" the engine knocking, it would certainly be more likely to survive such an episode with the booster in than without.

If, on the other hand, it is possible to alter the ignition advance curves in the ECU, then it may be better to use that approach. It's a difficult call to make because by altering the timing you'll put more heat into the exhaust valves, but you won't be spending an extre 3 to 5 pounds per tank for the booster. Conversely, if you did add the booster and the ECU were able to continue to run the factory advance, it is possible, but not not necessarily certain, that the exhaust valves would not get such a roasting (but, due to the slower burn rate, the effect may be to raise the gas temperature just before the blowdown phase and thus have similar, if not worse effects, than retarded timing... I'de need to measure EGTs in both cases to be sure... it may be quite cool in comparison but without measuring it I wouldn't like to say...)

Hugh,

It is unlikely that Optrand will make a closed loop ignition timing controller per se. I suspect that they will make the sensors and it's then up to the ECU manufacturers to add that functionality to their control code. Fortunately with their more ECU friendly 0-5V output it would be a simple case of sticking a high speed A/D on the ECU and a DSP, or to overspec the microcontroller and use spare cycles in that.

With regard to my comments about increasing power they had the following basis... your factory advance profile ran X degrees, but you had to knock it back to Y degrees because it was beginning to knock. As you have pulled timing away you are further away from MBT and thermal efficiency has fallen. Now by adding the "magic bullet" the timing can be increased again, but you will not necessarily make any more power because of the slower flamefront. Now contrast this to the situation with a "real" octane booster. By not altering (or altering to a lesser degree) the cylinder pressure curve, the peak in this instant is closer to MBT than it would be with the "magic bullet" version. This means that the thermal efficiency is greater at the same ignition advance, and hence the actual power output is greater at the same operating conditions, not because the fuel is able to release more energy (such as a racing fuel), but because the engine is able to more efficiently convert the thermal energy into kinetic energy.

Of course if you are not already at MBT and the octane booster is very good and allows more advance than the stock profile, additional advance will improve the situation further and even more power can be released, as you have stated. And if it's very good indeed then you might get a fuel which is has a higher autoignition temperature, and burns faster than the normal fuel; if the original timing was not at MBT, then this new fuel would make more power again at stock timing because by burning faster it is effectively moving the peak pressure closer to optimum without altering the ignition timing.

Cheers,

Pat.

Mycroft

Nitro, accepted.

Pat, nice posts!

Only one thing, the detonation that I believe still happens with OBs' will occur at a lower temp, we can see that with the coolent that is present in all of them, (by coolent I mean a residual light gas that buns cooler, this gas is present in the second det. and has a low flame progation though high ignition temp, similar to diesel) So the second detonation occurs but not so hot it therefore is not as immediately destructive as the high temp 'knock' we all know and hate, it is still destructive but is slower to form the piston head burning characteristic of the 'norm'.

Agree with the rest, 'cos it is correct!

I have got myself some Xylene R3 today and will try that at the weekend, should need about half the Iso-propyl (50cc) so should be fun.

Have you any input on the Digital sensors?
-------------------------------------------------------------
This thread (it seems to me) has become a little unwieldy, would anyone who has posted here object to the idea of pruning it and would anyone want to take on the task?

We have not finished but we are very much closer and i think anyone coming to it for the first time may just 'give up' and although a little'specialised' it has now become essential reading, to clutter it less, please post your answers to this (and we do need consensus) on the thread I will start momentarily.

somberg

Hugh,

You're right re. the need for recalibration of these under-spark plug sensors. A bit user-unfriendly that is.
The MEAA Ionic Current Detection looks very interesting. The information provided on that web site is quite basic. Is there some more info somewhere re. these or comparable products that you know of ?

Re. Motec: they say it's possible with the M800/M880 to use an analogue input to trim the ignition timing. I'd prefer this to be implemented using CANbus messages since I'd be able to connect a black box with an algorithm inside to control ignition advance with individual cylinder (knock) pressure sensors + others (like crank angle) as inputs using the CANbus to output control data. Since I'm not yet familiar with the Motec hardware, I can't say if it's possible using the CANbus.

To be continued.

Andre.

__________________
Bananaman..

somberg

How about this:

J. Auzins, H. Johansson, and J. Nytomt. Ion-gap sense in misfire detection, knock and engine control. SAE Technical Paper Series, 950004, 1995.

Robert L. Anderson. In-cylinder measurements of combustion characteristics using ionization sensors. 2SAE, 860485, 1986.

Fei An, Giorgio Rizzoni, and Devesh Upadhyay. Combustion diagnostics in methane-fueled SI engine using the spark plug as an ionization probe. SAE Technical Paper Series, 970033, 1997.

K.N.C. Bray and N Collings. Ionization sensors for internal combustion engine diagnostics. Endeavour, New Series, 15(1):10-12, 1991.

Some work to do on a rainy Sunday afternoon...

__________________
Bananaman..

pat

Mycroft,

you may well be right with regard to a lower level knock occuring with the coolent (sp?), but I have found no evidence of this on any piston crowns I have inspected, where engines have been mapped and run with OBs. It may be that although it is still present, its intensity is reduced to such an extent that it is not destructive anymore? Perhaps corresponds to a praticular artifact that sounds like a "fart" through det cans... that certainly is an abnormal combustion even but doesn't sound like someone dropping a pin in a tin can.... but then I may be confusing that with the sound of preignition... too many different types of abnormal combustion modes to remember... LOL

Good luck with the Xylene... at least this one's not carcinogenic...

With regard to the digital knock sensors, I'de have to do a bit of digging...

Cheers,

Pat.

somberg

How about this:

Knock and misfire monitor

With a sensor the output noise of the engine is measured during each rotation of the crankshaft. With a fast microprocessor the acceleration and deceleration of each piston movement is tracked. The difference of the learned default values and the real-time values are interpreted.

With two different outputs the engine correction for knock and misfire are given.

Andre.

__________________
Bananaman..

somberg

Hugh

Have you seen this:

http://www.delphi.com/pdf/eandepdfs/Ignition/IonCurrentSens.pdf

Andre.

__________________
Bananaman..

Hugh Keir

Andre,

Hadn't seen the Delpi stuff before, but I have seen a controller chip somewhere that has been produced as a stand alone device to provide the management of the per cylinder ignition timing and might with a bit of work integrate with your Motec.

Back home for a day and have dug out the Hyundai paper, SAE 2000-01-0933. Closed-Loop control of Spark advance and Air-Fuel ratio in Si engines Using Cylinder Pressure.

They used Kistler spark plug pressure sensors which were used to measure pressure every degree of crank rotation.

The main control seemed to be based on the expected performance at 5 points from 40° Before TDC to 40° after TDC. Using this strategy, they were able to monitor the burn process and hold peak cylinder pressure at 16° after TDC for performance and found 38° advance worked for lean burn cruise at 3000 rpm.

This provided knock, air / fuel ratio and ignition control for optimum performance.

As far as I am aware SAAB have definitely gone further than most on this as I believe all their cars use ion sensing as standard.
Will try and find out about the control chip.

You can look at www.adrenalineresearch.com/smart.htm and see a system that works and you can buy. Prices start at US $22,475.

There is also a DIY system that has been started. see www.hut.fi/~vvartiov/ion/ion.html

Gota go my missus is bending my lug

somberg

Hugh

Thanks.
I found that Adrenaline stuff after I posted you; looks quite good; bit expensive though...

I also found an enormous amount of hits on Swedish university web sites; I agree in thinking that they are quite far with this in Sweden (Saab). They're also using the principle on diesel engines.

Looking forward to your data on that chip !

And good luck with the missus..

Andre.

EDIT: the 1999 Volvo S70 uses it, BMW also apparently. Wouldn't be surprised to see it in the next gen Nissan as well (like 350Z, R35).

__________________
Bananaman..

nitro

MYCROFT what is your email address and ill send you a chart of the independent test and the results including your toulene from a knock engine.
These test were carried out under the ASYM D 2699

__________________
When only the BEST will do!!

Mycroft

It's in my profile!

So to summarize, the 'Octane Booster' seems to work best by allowing you to advance the timing, but this can be a bit 'hit and miss', if you don't advance your timing then you only get the original settings maintained for longer.

Is that fair?

If so then the next question that screams for attention is whether you should now look to run the fuel with the highest Calorific Value. This is NOT Optimax, the fuel company that produces the fuel wih the highest CV is the Total/Fina/Elf group SUL, strangely this also uses Manganese to replace TEL and Manganese also smoothes the Benzene firing.

the TFE group fuel has a 5.5% higher CV than it's nearest rival, (Optimax), so if you are adding (to some extent) the Aromatics found in Optimax already then surely we must look to use the highest CV fuel for the base!

Whaddya think?

Hugh Keir

Mycroft,

Spoke to the chemist at Silkolene a year or so back about octane booster. He said that the product that had the greatest effect on octane was iron - better than lead, problem was it screwed up the internals of your engine.

I agree calorific value against octane would be great.

I asked Shell for details of their fuel some time ago and all I got was a generic sheet with no specific details.

Can you give us some more info on the various fuels?