Technical cooling number 4

[fingers99]

Originally Posted by fingers99 (somewhere else)
Today I started preparing to pull the head off Blue. And I pulled the plugs. Number 4 has both electrodes molten (classic sign of detonation, when I pull the head I'm sure number 4 piston will have a hole in it). (Pics in my blog before I go to bed.) All the other plugs are fine. Now, as Aaron predicted, it was number 4 that went and that seems to be the problem cylinder on SPI turbos. Fuel distribution may well have something to do with it. Blue is going MPI.



Pic of ye plugs. The rest are remarkably similar, despite what the pic shows.

Just as I thought, number 4, its a design flaw on the inlet manifold side that I had hoped would be better with the UNO inlet as it has nicer design, and IIRC it runs hotter than eth other ports due to the way the water ciruclates the head. I think and this is only a theory that it is to do with the brake servo vacuum coming from the inlet manifold on that side, though TBH 1 and 4 being the two furthest away from eth throttle body are probably always going to have marginal fuelling compared to 2 & 3 being in the middle.

I found some ideas in Bell (where else?) on modifying the block to increase water flow to the “rear” (that is, cylinder 4) of the head. The problem of number 4 overheating is likely to occur on the 1242 engine as well, but might be less of an issue on the 4 valve engines which not only have more valve circumference than the 8 valve ones (which should aid cooling for the exhaust valves) but have central plugs, which might see more water anyway.

The easiest modification is simply to radically reduce the size of the water passages at the water pump end of the head at the interface with the head gasket, down to somewhere between 1/16 to 1/8 inch, either by plugging with silicon or epoxy or an alloy plug. This means that more water flow is directed toward number 4 where it is needed. Of course, the holes still have to line up with the holes in the head gasket.

Moving on from there, there are possible modifications to the water pump. Until I've held the new water pump for Blue in my hands I'm not sure whether it will warrant modification (or even if modification is possible), but everyone out there should be wary of fitting a pattern pump to a tuned engine: some of the impellers I've seen are very poorly designed – obviously made for ease of production rather than efficiency.

Next step (I won't be doing this!) is a separate pump with manifolding designed to direct the water exactly where you want it and allowing you to (for example) direct 25% of the flow to the block and 75% to the head.

Next step from there (and only applicable to 4 valve engines) is to do what Lancia did on the Delta S4 and arrange the valves so that, instead of having all the inlet valves on one side of the block and all the exhausts on the other, they ran inlet/exhaust/inlet/exhaust/inlet/exhaust/inlet/exhaust. But that would require radical reworking of the head (probably a new casting) and a complete redesign of the camshafts........... Still, if you need to produce 450 – 600 bhp from a 1.8 litre engine and have it perform reliably on rally and road stages, what else can you do?

 

[allanhelen]

I am absolutly no expert in head design by any means but this rasies a few questions in my mind reading this.

How quickly is the critical temperature reached is it rapid or "heat soak"?

Is the head able to conduct the heat away fast enough even with better water flow? what if the critical temp is reached rapidly surely there is a limit to the conductive properties?

If you direct more water flow to No 4 does this compromise the integrity of the other 3 cylinders?

Could you not replace the Current water pump with a higher flowing one?

What about water or methanol injection?

Would avoiding extended periods of WOT throttle stop this happening

 

[Brooky]

about the water pumps...which are better,the plastic or metal impellors??

 

[custard]

would an electric pump which isnt limited by revs be a practical solution?

 

[fingers99]

How quickly is the critical temperature reached is it rapid or "heat soak"?

I'd imagine that it's pretty quick given WTO. Detonation will do real damage in nanoseconds.

Is the head able to conduct the heat away fast enough even with better water flow? what if the critical temp is reached rapidly surely there is a limit to the conductive properties?

I think that the cooling is probably OK given that the bulk temperature was fine. There will be a point at which it runs away, but aluminium is a great conductor (as is water).

If you direct more water flow to No 4 does this compromise the integrity of the other 3 cylinders?

Seems not. The trick is to even it out. The dynamics of the flow seem to be very similar to those you get in the inlet manifolds of charged engines, where the cylinder on the end doesn't get its fair shair of air/petrol.

Could you not replace the Current water pump with a higher flowing one?

I guess you could if someone made one, but you can easily bump into issues like cavitation, and the water flow would still be uneven.

What about water or methanol injection?

Just something else to go wrong. It'd work, but it ain't cheap, and where to put it? Hopefully my EGT sensor will catch it (if it ever rears it's ugly head again).

Would avoiding extended periods of WOT throttle stop this happening

If it did, it'd point to a a serious underlying issue which needed to be addressed, wouldn't it? And, I was only asking for WOT when it happened. I think you could well run into similar problems on a drag strip. Like I say, detonation does damage very quickly.

 

[fingers99]

about the water pumps...which are better,the plastic or metal impellors??

It's not really about material, but design. Some of the metal ones I've seen look like someone's taken a pair of snips to a big washer and turned the edges up. So, I'd think the plastic ones stand a fair chance of being better.

 

[fingers99]

would an electric pump which isnt limited by revs be a practical solution?

I doubt it. You may be increasing the mass flow, but might run into cavitation. If I understand it correctly, folk are fitting the electric ones so they minimise the flow rate (and thus pick up a bit of HP). It'd probably be the only solution if you wanted to split the flow between block and head, though.

 

[Brooky]

Seems not. The trick is to even it out. The dynamics of the flow seem to be very similar to those you get in the inlet manifolds of charged engines, where the cylinder on the end doesn't get its fair shair of air/petrol.

I have a solution.....but i wont say as i know you dont approve of it

 

[fingers99]

I have a solution.....but i wont say as i know you dont approve of it

But we needs horsepower and we needs torques! It's all down to the shape of the plenum, missus!

 

[arc]

i think cooling the inlet charge will be a more cost effective, and productive approach to this problem.

 

[Brooky]

i think cooling the inlet charge will be a more cost effective, and productive approach to this problem.

Nitrous oxide:slayer::slayer::slayer::slayer:

 

[fingers99]

i think cooling the inlet charge will be a more cost effective, and productive approach to this problem.

It'll help. But cost effective? Compare price of a bit of scrap aluminium and some araldite to the cost of water injection/bigger intercooler.

Of the two, the bigger intercooler might increase performance and raise the point at which it happened, but water injection is pretty awkward and not that practical for road use (or on something as teeny as a Cinq).

Interesting pics in Bell of what BMW did to convert the old 2002 block to take (what, 1000 bhp?). Essentially they just modified the water passages.

 

[allanhelen]

I supose the secret is to find out where and by how much to modify the waterways to give the desired flow charicteristics the only way you'll know for sure without some serious technical kit is if it doesn't hole a piston.

At least cooling the fuel air charge you have a chance as a DIY person of measuring this. If you've got an SPI as well chances are you have an MF2 and isn't water injection what they were origionally designed for? (i need to do some reading up on these things)

the other option again of course is to try Aarons hunch and reposition the Servo vac line or do away with it completly.

Another one off the top of my head, would it be practical to flow the manifold so that it slightly favoured 1 and 4 to redress the normal 2 3 favouring?

 

[fingers99]

Dunno what the accursed MF2 was originally designed for. It could be set up (with a pump and methanol resistant tank and lines) to give water on boost, but that's a half arsed way of doing it: what you really need is some way of tying it to EGT, knock or even inlet temperature. As I say, I don't think water injection is very practical for road use.

The inlet manifold mod you suggest is possible, but very difficult without a flow bench. The books give a plenum design which will work better, but that will be evo 2 for me.

Essentially, there are two different problems here which converge on detonation. One concerns the inlet temperature, the other the ability of the coolant to pull away heat. To get really high outputs (say 160bhp + from an 1100) chances are you'd need to address both. But first you'd need something to control the torque steer!

 

[J333EVO]

Another thing just to confuse further.

The breather system on Cento's is not brilliant, though the addition of a catch tank helps, but hot oil vapour has a habit of igniting and causing pre-ignition.

It could be an area to look at, look how it is on turbo cars as standard.

But with MPi, and cold in take temps, i wouldn't go too deep into trying to solve issues, just the fact its MPi with injectors close to combustion chamber adds cooling to the engine.

As for water ways, when you sit a 16V gasket over an 8V gasket, the way it has openings for the water is a bit different, larger on some much much smaller on others, this is to control the flow through the block and head optimising it on the otter bits.

 

[Brooky]

The breather system on Cento's is not brilliant, though the addition of a catch tank helps, but hot oil vapour has a habit of igniting and causing pre-ignition.

Would a crank breather from the sump be any good also helping to relieve any pressure??

 

[J333EVO]

Would a crank breather from the sump be any good also helping to relieve any pressure??

Yes, Emma's old car (Martin's) has a sump breath on it which we believe can only make it breath better.

 

[fingers99]

Certainly true that oil vapour effectively leans the mixture out, thus causing detonation. If I can find the right bits (just hope I didn't chuck em out!), I'll put together a sump breather.

MPI can only help with the fuel distribution. At some point I've got a nice old alloy 75 manifold which I intend to cut up, but, as I say, that'll be evo 2.

As regards the water flow, I'll be cautious. Great big 1/8" holes at the #1 end and it's less than a day's work. Plug and piston top colours should give me some idea of what's going on before I go more radical, and with luck I'll get it right enough first time.

 

[Brooky]

i have a spare alloy inlet if it all goes wrong

 

[allanhelen]

The inlet manifold mod you suggest is possible, but very difficult without a flow bench. The books give a plenum design which will work better, but that will be evo 2 for me.

hmmm now one of the things i've always thought about the SPI bob design was that it looks like and after thought there must be a better way. After all even the second injector isn't even in a consitant possition you would have thought the possition of this might be important?

edit sorry bit of a delay writing and posting thanks to the joys of air travel

my whole participation in this post, selfishly is to protect and preserve my own current engine. As it is high output as it is I do not want to explore an envolope that has been well and truely expanded, and no doubt will be taken further by others and burst:worship:. I personally have already been there and paid off the overdraft on other projects.

What i really want and i suspect others too want to know is there anything more i can achievable do to keep my engine in one piece without driving like an 80 year old granny?

edit, edit

Yes, Emma's old car (Martin's) has a sump breath on it which we believe can only make it breath better.

And you would be suprised at the amount of water vapour that it kicks out
with these lovely cold temps, at -4 outside temp the other day i actually could see the catch tank filling up!