[engine_work] Project Sarty

[DrSarty 90 1 Cars]

I would guess he'll be running wasted spark because the Mi16 head doesn't have any camsensor as standard, hence he can't run coil on plug as it is. Could probably be fixed by using a S16 thermostat housing with camsensor etc., but as long wasted spark works well, why bother.

 

^^^ What he said!

 

 

Here's a diagram I doodled which I hope reflects PeterT's explanation. I have two questions on this then:

 

1) Could someone please check I've interpreted the connectivity correctly?

 

& secondly

 

2) Could it literally be wired like this, with the earth/ground source being through the strut brace rather than individual wires tracking back to another earth bolt?

 

Oooh, oooh...and PPS, where can I get the connectors x 4 to plug into the coil packs & make up my loom?

Edited December 19, 2007 by DrSarty

[Super Josh 4]

Rich, I've on thing to say..... Matrix Reloaded

 

 

 

Josh

[GLPoomobile 958]

Nice one Rich, can't wait to see the outcome of this setup.

 

It's almost like a whole forum project with the amount of involvement from various members that you've managed to secure Sorry I can't contribute........unless you need someone to stand around and get in the way, maybe occasionally break something for you

[Henry 1.9GTi 36]

i see

 

would using 1 of the LS1 coils per 2 cylinders produce enough spark energy for a sufficient burn? Means 2 less coils. Half price! which is always good in my book.

 

i see the idea though which is a good one as the wasted spark isnt a problem and it means you can run the megajolt ECU without having to fork out loads for an ECU that can control each spark induvidually.

 

good thread.

 

Henry.

[vern 0]

Mr Glpoo could always offer to stay in and wait for any parcels for you, as he is now on first name terms with most if not all, the parcel delivery companys in the country. He always recieves excellent service and all parcels should arrive before they are sent.

[GLPoomobile 958]

Mr Glpoo could always offer to stay in and wait for any parcels for you, as he is now on first name terms with most if not all, the parcel delivery companys in the country. He always recieves excellent service and all parcels should arrive before they are sent.

 

 

Errrrr, do you want Rich to get this project finished or not? Cos I don't think it is likely to happen this century if I was responsible for handling deliveries

 

EDIT: Rich and I were discussing this yesterday and I said I pity anyone who moves to my flat after I leave, as it's probably blacklisted by every delivery company now He said they'd be alright as the companies will just track me to my next address, but I pointed out that they can't find my current f***ing address so it's pretty unlikely

Edited December 19, 2007 by GLPoomobile

[petert 549]

It is, in effect, coil on plug, or direct fire, as there are four individual coils, one per cylinder. The fact that they'll also fire on the exhaust stroke is irrelevant!

 

Yes, Rich, your diagram is correct. I'd still run an earth wire, rather than relying on contact through mating parts.

[Henry 1.9GTi 36]

yes your right ofcourse peter, but because HT leads are still being used it seems pointless having 4 coils, instead of just the usual 2 for the wasted spark system. But at the end of the day, it works and may be cheaper than trying to find a suitable pair of coils.

 

Henry.

[DrSarty 90 1 Cars]

yes your right ofcourse peter, but because HT leads are still being used it seems pointless having 4 coils, instead of just the usual 2 for the wasted spark system. But at the end of the day, it works and may be cheaper than trying to find a suitable pair of coils.

 

Henry.

 

Henry, now that to me isn't making sense.

 

Even with wasted spark from a typical coil pack you have 4 off pretty standard HT leads between the coil pack (which literally is a 'pack' of two coils - whereas as my set up is really like Pete said 'coil over plug') and the spark plugs don't you???

 

The ECU connects to the coil pack, and the coil pack to the plugs = 4 HT leads. My set up will have a dedicated coil located on the strut brace directly over the spark plug it fires, via a very short (and fat ) HT lead. The coils x 4 will - as Pete's explanation shows - be seperate items but wired into 2 pairs, so will be like a typical coil pack that's been broken in two, and this pair will fire 1 & 4 together and then 2 & 3. On each fire, either 1 or 4, or 2 or 3 will be on their power stroke, so one of the sparks will be igniting a mixture and the other will be wasted (man ).

 

Yes I'll have 4 coils rather than 2, but I'll have less HT lead length although I'll still have 4 HT leads just like most other set-ups. It's probably an exercise in vanity and being distinctive rather than anything else.

Edited December 20, 2007 by DrSarty

[petert 549]

Two plugs on one coil is much less effective as the total energy of the coil is distributed over two plug gaps, not one. The way Rich is doing it gives all the advantages of direct fire without the hassles of having a cam sensor (or other reference signal).

[welshpug 1,621]

what you're proposing is near identical to the setup used on the 2.0 Mi and S16, i.e 4 coils in a wasted spark system.

[DrSarty 90 1 Cars]

what you're proposing is near identical to the setup used on the 2.0 Mi and S16, i.e 4 coils in a wasted spark system.

 

This is very true. I suspect it is very difficult, nigh impossible to come up with something unique which hasn't been done with one of these engine variants.

 

It also poses the point that Peugeot must themselves have decided to go this route for a reason, and I saw this set-up on PeteT's 2.0 race engine, hence me going for it too.

 

I'm just hoping Matt @ QEP has taken some pictures for me/us of the build. I did ask him too.

Edited December 21, 2007 by DrSarty

[Super Josh 4]

what you're proposing is near identical to the setup used on the 2.0 Mi and S16, i.e 4 coils in a wasted spark system.

 

 

It's not though, is it? The S16 has a cam phase sensor and uses sequential injection and not wasted spark either.

 

 

 

Josh

[petert 549]

The S16 is sequential injection and wasted spark ignition. There are only two ignition outputs from the ECU, which trigger two ignition modules. If it were a true direct fire setup there would be four dedicated outputs, triggering four modules.

[welshpug 1,621]

The S16 is sequential injection and wasted spark ignition. There are only two ignition outputs from the ECU, which trigger two ignition modules. If it were a true direct fire setup there would be four dedicated outputs, triggering four modules.

 

like the gti6 IIRC Peter.

[DrSarty 90 1 Cars]

For the benefit of the thread and those not too sure on all this coil pack nonsense, here are a few pictures to show a typical coil pack (for 4cyl engines) and my new individual coils. Coil on plug BTW is literally a coil (pack) that plugs directly onto the spark plug itself.

 

A COIL PACK:

 

3 PICS OF MY ***AC DELCO 12558948*** COILS

 

As you can see, they're not particularly big. They are quite weighty though as that shiny loop thing you can see (with 2 holes in) is a multi layered metal ring and (I guess) core for the coil itself. The main body is plastic and includes the recepticle for the HT lead going to the plug. The 4 pin connector just like PeteT's diagram ref wiring is shown above the HT socket. They come individually boxed and in this instance came from Fresno, California.

 

PeteT: Please could you assure me by way of recognition or checking the part no. that I have the correct items please? They are stamped into the plastic at one end: 12558948 & 06H13-0200, which I'm guessing are part numbers and batch nos. respectively. The box states: GM# 12558948 & ACD# D580.

 

Here's hoping I haven't made a moron out of myself and got the wrong items.

[Sandy 189]

I don't have experience of individual coils, so I'm sticking my neck out a bit by saying this, but what's wrong with a £10 secondhand wasted spark coilpack like the one shown above? If some of the engine builders I work with are seeing over 140bhp/litre and 10,000 rpm limits with them?

[Henry 1.9GTi 36]

i only said the above because usually with 4 coils they are coil directly onto the spark plug like you said, but theres no HT lead. All the high tension side of things is directly into the plug rather than down a long lead. Obviosuly the Mi16 wasnt designed for this so you probably cant get a coil to fit down there? Where as with the normal wasted spark HT leads are retained like your doing. No way is wrong as such, just different ways of doing things, so just chatting it out, not telling you how to do it as i have no idea which way would be best for your requirements

 

Neway my knowledge is limited but its all very interesting

 

keep us updated!

 

Henry.

Edited December 23, 2007 by Henry 1.9GTi

[DrSarty 90 1 Cars]

I agree with Henry, Sandy and Pete on this: i.e. there is no specific method for doing this.

 

Sandy is right in so much as I've almost over complicated things, and definately defeated one of my 'low cost' objectives by spending £110 instead of £10. PeteT is technically guilty in this matter as it's what he did; and I, rather unusually, just copied him. I did explain earlier that I did it to a certain degree for vanity reasons .

 

I was originally going the EDIS route using a new trigger wheel and a coil pack, which is what started me having to learn about this. Ironically, once I spotted Pete's 2.0 race engine with the individual coils and chose that ignition route, I learned that the S16 or '6' have a similar set-up. In effect I'm admitting I could've spent £100 less: but I wanted to be adventurous and different.

 

As for 140bhp per litre I'm staggered . A friend of mine disagreed that my engine 'as is' would/could make 200+, as that was unlikely without VVT (variable valve timing) etc as it was needed to break the 100bhp per litre benchmark. Sandy's comment clearly trashes that, ALTHOUGH I would say I figure this is acquired through the increasing trend of building higher and higher revving engines. Sandy's recent 1.4 etc and others he's referring to revving to 10k, is similar to Audi RS4 and BMW M3 V8s revving to 8k!!! One of my objectives was to get the torque without the drama, which kind of bucks this trend. It'll still come alive at 4.5-5k I suspect by nature of the design of the Mi16 head and the cam 'coming on song' at that point. But I doubt it'll be an engine revving much past 6 or 7k IMHO.

 

But I do like to be different.

 

And Sandy: I'm down in Okehampton (Devon) for xmas & new year. So if you fancy a meet-up half way-ish for that bevvy - and anyone else BTW - we could arrange it quite easily.

Edited December 24, 2007 by DrSarty

[Sandy 189]

Sadly I haven't managed 140bhp/litre yet, I was referring to John Read, Mark Shillaber and Pete Willis, who are regularly getting that sort of specific output, or even better from several engine types. My record so far is only 129bhp/litre, but I really haven't built that many engines! I tend to make mid range power the priority, getting good torque over a wide usable range, which is what makes a car easy to drive quickly. Most of my customers are more enthusiast, than driving god and this is usually more suitable!

 

What I would encourage you to do, is chase a substantial torque curve, which frankly takes more skill and better matching of components than simply building an engine for revs! It's perfectly possible and reasonable to expect 85-90lbft/litre from a well rounded XU 16v on hydraulic lifters. I think what probably thwarts that in many cases is having one element too zealous (too much cam duration, inlet ports too big or inlet valves too big etc), or more commonly, not paying sufficient atention to the inlet and/or exhaust geometry. I prepare to be flamed

[DrSarty 90 1 Cars]

Ironically I am an enthusiastic driving God!

 

Reference 'flaming': it's very intersting to see your comments on valves/ports actually being too big and restricting performance.

 

Were a certain forum member named after a type of cat still participating on here, I'm sure he'd have something to patronize us with - I mean explain to us.

 

And the meet is arranged.

[kyepan 291]

sandy, what are the rules for inlet angle? is there anything we can read up on to understand how it affects torque?

[kyepan 291]

something like this but for angle

 

and this is nicked from here

It’s possible to obtain greater than 100% volumetric efficiency in a naturally aspirated (non-supercharged) engine by using tuned intake/exhaust systems. How all this works isn’t all that complicated, but it takes a while to explain it properly. I think the first use of tuned intake runners were the vertical velocity stacks. They were generally of a length that put the open bell mouth somewhere around 15 to 18 inches above the back side of the intake valve. Although both the length and diameter of the runner are important, the length is what determines the specific rpm at which the runner is tuned to provide peak efficiency.

 

The velocity stack utilized the fact that air is a compressible fluid to produce its boost. At wide open throttle, with the engine turning at high rpm (the rpm that the intake is tuned for), a column of air is moving at a high rate of speed down the intake pipe while the intake valve is open. The column of air moves toward the cylinder in response to differential pressure; the pressure in the cylinder is lower than the pressure at the open end of the velocity stack. When the intake valve closes, the inertia of the column of air causes it to continue moving down the intake tube, stacking up against the back of the closed intake valve, causing the intake air to compress, and creating a higher pressure right up against the valve.

 

Now if the high pressure air sitting at the intake valve would just stay there until the valve opens again, life would be very simple. Only it can’t, because the pressure is now lower at the inlet to the runner. So the high pressure air bounces off the closed valve and tries to move backwards toward the inlet. Since the valve is closed, the entire column of doesn’t really start flowing backwards; instead it is more like a high pressure wave propagating back toward the inlet. This high pressure wave (or pulse) leaves a low pressure behind it, and when it finally reaches the inlet, the pressure at the inlet is now greater than the pressure in the intake tube. As a result of low pressure in the tube, air starts moving back into the tube, its inertia causing it to stack up against the intake valve again, which is still closed. If the engine is turning the proper rpm (whatever the intake tube is tuned for), the intake valve opens when this higher than ambient pressure is present at the valve.

 

Some of the design considerations are pretty obvious. If the diameter of the tube is too large, the velocity of the column of air will be too slow to create a good inertial pulse, or reflected wave. If the tube is too narrow it will restrict the airflow and cause a performance decrease. The length of the runner determines the rpm where any boost effect will occur. Earlier it was noted that the typical velocity stacks on old race cars were around a foot long. Since I think the reflected waves that set up inside the runner propagate at the speed of sound, I think the length is such that it is three times as long as the calculations would indicate it should be. These older types used a third order harmonic, or in other words, several of these waves would be bouncing back and forth inside there at any given time. Maybe some math people here can sort out whether this is part is correct or not. I think they used to use a foot (or foot and a half) long for an rpm around 5 or 6k. It seems like there used to be a formula that was used to make some sort of preliminary “length to rpm” calculation. Testing would still be needed to fine-tune a particular setup.

 

The way it was explained to me, using a length tuned to the third-order harmonic gives a very deep peak when you hit the resonate rpm, in other words a big kick. The problem with this is that it is not effective when you get off of (above or below) that rpm for which it is tuned. The much shorter stacks that are common today, and the tuned induction systems seen on a lot of cars must be using first- or second-order harmonics, I’m not sure, but that would have the purpose of making the thing effective over a significantly wider rpm band, but at the cost of not producing quite as high of a peak boost.

 

Exhaust systems are similar, but I think they are not quite as twitchy to get right as the intake runners are. The exhaust pulses have a significantly higher pressure differential, or power pulse, that you are dealing with, to begin with. But the process is similar, except the thing is tuned to have the exhaust valve open when the low pressure pulse is present at the back of the valve.

dad tried to explain this whole pulse tuning thing to me years ago, i never understood it until now.

Edited December 25, 2007 by kyepan

[DrSarty 90 1 Cars]

That's fascinating stuff Justin, and what you were alluding to over our recent naan breads.

 

No really; you seem to have stumbled upon this at the same time as my mate Guy (who commented on needing VVT technology to reach these staggering BHP per litre figures) when I asked about why a cam 'comes on song' so to speak: i.e. what's happening in my Mi at 4.5k rpm? Of course this applies to all other N/A engines that acquire a new lease of life at x rpm.

 

Even Guy understood it was about harmonics and balancing air flow, which is how he explained the 'on cam' event, when these flows effectively get into synch in the inlet-head-exhaust cycle. And this has also now shed some light on the GTi6 inlet (the one I'm using) having those plastic resonance chambers. However I've since learned that even that would've been tuned in with the '6' exhaust system which I don't have, let alone the head!!!

 

At this point it is fair to mention that I now hold an even higher regard for those knowledgable in the internal combustion engine, aka proper automotive engineers.

 

What I mean is that all this big valve stuff sounds good, but as I've learnt, in cars and everything in life, there are always pros and cons to everything. Hence bigger is not always better with regards to valves. It may sound good and sell easily to get more air and fuel in by way of a bigger valve, but as Sandy is subtly hinting at, depending on your design goal i.e. peaky power for use on a track or tractable broadband power for flexible road (and perhaps rally) use, it takes great skill to match all of the flow rates - in & out - as well as all of the components to create a truly great engine.

 

This should spell out now that I'm lobbing a few OK components together and just hoping for the best. The real skill with my engine (perhaps after a bit of luck with combining these components) will be in the set-up and mapping. In other words I'm taking a bit of a gamble here. It may pay off; I'm lucky like that. It's a bit like Chelsea hooning loads of cash out several seasons ago on the 'best' players - as dictated by their reputations & transfer fees - and then watching it not work for several seasons until the 'components' actually started to work together.

 

Another thing Justin (Kyepan) was telling me about, was how obvious it is really that mechanically the engine components should be matched and balanced too. A good analogy would be trying to push a shopping trolley really fast and straight, when it has (inevitibly) a wobbly wheel/castor. So much of your effort i.e. energy is going to be wasted in its innefficiencies. Get the wheel sorted out first to run true, and more of your effort can be used to drive the trolley forward. Hence a crank shaft, flywheel and piston balance, although a little costly will undoubtedly either eek more power out of an unbalanced engine, without adding any parts, or at the very least, gain you more benefit from any future tuning and better return for your time and money.

 

Interesting stuff I think.

Edited December 25, 2007 by DrSarty

[Sandy 189]

Absolutely.

 

I always insist the point as well, that it's near impossible to accurately model what the geometries "should" be, although some claim to be able to. If that was so, then all competition engines would surely be built to the same formula. The problem is, you have changing sonic, fluid dynamic and pressure events occurring simultaneously and they overlap and influence each other dramatically. It's closer to chaos than order. The most reliable way to achieve good results IMO, is to experiment, collect the data, find the trends and work from them with an open mind, without bogging yourself down with theories and models of what should be happening.

Not many people on here really have the budgets or facilities needed to experiment very widely, so my thinking is choose moderate engine internals and play with the externals to improve their efficiency.