[Mk2] [Turbo] High comp low blow anyone?

[Loser]

i'm going to go through that post here as i read it again- i have read it before but i want to read it again here because i don't fully understand why it makes my idea a bad one.. Sorry for being dumb

i read on here that someone using an st205 chargecooler never saw intake temps higher than 35 degrees C at 15 psi on a ct20, so i'll use that 35 degrees C, even though i won't be compressing and therefore heating the intake charge that much it'll be from a less efficiant turbo so all being well it'll be approximately that.

If i just straight plug the numbers in i see a charge temp of 580 deg F (assuming a dynamic ratio of 9:1 (static being 10?) and i belive the flashpoint of air fuel is 600 deg F, but i read that on an american site and i'm unsure of what octane that fuel was. Is that too close to that limit? I'm also unsure if the random numbers in that equation are related to engines in general or the 3sgte. And presumably it won't take into account heat soak from the head and cylinder?

With me reducing the boost level won't i be in turn reducing the backpressure ratio of the turbo? I'm not sure of the extent of that, but at 15 psi through a cheap ebay smic it'll be putting up alot more back pressure than in the proposed system, in fact beyond 5.5k it can't even hold 15 psi- the boost level drops to around 12 before redline, so the wastegate would be fully closed!

I'm totally stumped by that 3rd graph, can't visulise how it works.. I understand that the 2d graphs can be combined, but not understand how the graph works.

Do you know if anyone's mapped the back pressure ratio for a ct26?

I'm understanding alot better now actually, i'm off to go learn more about intake air temperatures, hopefully the backpressure ratio of a ct26, and i'm still puzzled as to why a honda vtec can be mapped at 12lbs with it's compression ratio of 10.4:1 - the ones i've seen didn't have massive intercoolers, and i shouldn't have thought they'd be much if at all superior to an st205 cc...

I like learning

[Loser]

just been doing some more noseying around and managed to find someone who managed a 70 deg C inlet temp- running 17 psi on a ct20, assuming that's running a standard compression ratio (i will pm him) then that suggests a 633 deg F tdc temp- and apparently it didn't blow up

how does that work? (sorry all, feeble mind at work here) that works out a higher tdc temp than me, but i'm still unsure on the exhaust back pressure. Not looked into that yet.

[Wingers]

I've never tried to estimate in-cylinder temps based on comp ratio but I will ask someone at work as that's exactly the kind of thing that they do on a daily basis. This could give some idea of the likely change in temps based on CR but working out what the max allowable temp is would still be virtually impossible. A poorly designed combustion chamber will create hot-spots that promote pre-ignition/auto-ignition so that affects how hot you can allow the charge to get. As well as that, ignition timing also has a big effect so, for example you may find that the ct20 running 17psi mentioned above was either running some kind of knock resistance (e.g. water/meth injection), or was running a very conservative ignition map to get away with it. Hence killing the power anyway.

Understanding what the trends are is one thing but working out what values you can really get away with is a long task that normally needs a lot of testing.
I did think about creating a WAVE model of the 3sgte at one point but it seemed pointless without taking lots of measurements needed to calibrate the model. Realistically, I think other peoples test results may be your best starting point to doing this, but I don't know if anyone has really tried this before? It sounds like you've been doing plenty of research so you're probably as well placed as anyone to answer that question.

Wingers

[Loser]

i'm wondering how much effect heatsoak would have on the intake charge temps at tdc- simply because use charge will be moving so quickly- off the top of my head 200 mph+ and only some of the charge will be heated- where it's in contact with the head/block.. Is this why you see higher compression na cars the higher they rev? Digression time lol.

Please report back what the fellas at work have to say

[Wingers]

OK, I’ve had a work with one of the experts at work and he pretty much agreed with what’s already been said. He recommended that it’s not generally a good idea to trade boost for CR as you end up taking out a lot of MAP to compensate for a small change in CR. Also, sensible targets for CR are often taken from benchmarking existing engine designs as it’s very hard to predict knock limits. He also said that you normally define the performance you want, then increase the CR to the point at which you have to retard timing and/or use significant fuel cooling to protect from knock. (This is where I think you might be going with your plans but the CR change could be minimal)

Some key points to remember are:

Typically 1CR = 2-4% fuel economy gain.

In cylinder temps are one of the main drivers for knock limit.

Combustion chamber design also affects knock but since we’re dealing with the same design for both set-ups then we can (roughly) estimate knock limit by working out peak in-cylinder temps.

Combustion temps increase with CR from extra charge compression, but also because the combustion process occurs faster with higher CR




Based on that, it sounds like estimate peak temps using gas laws is worth doing. Without checking the numbers, I think that’s what you’ve already done but remember that the process isn’t isentropic so the real temps will be higher than you’re predicting. This will be the case for both low & high CR variants but probably more so for the high CR engine as it’s doing more work, hence will be further from an “ideal process”. Also, the peak temps for the high CR engine will be worse still due to the combustion occurring faster. Unfortunately I don’t have a grasp on how much worse so for now we’ll have to leave this out of the calcs and just bear it in mind.

The trouble with your example is that (unless you know otherwise) there is nothing to say that the engine is running maximum possible boost. I’ve heard of 2bar boost or more, but I haven’t seen that first hand so don’t know how effective that is, or what charge temps it’s combined with.

I think the best thing to do would be to identify an engine with a known CR, IAT and MAP that’s running on the knock limit.
It would then be possible to use that to setup a baseline and compare its peak temps to a high CR version of the same engine.
The important thing is to consider “knock limit” as the point at which ignition/fuelling needs to be used to protect from knock, at the expense of power. That means that the engine can’t be one running a huge amount of boost and using the ignition map/fuelling to survive.

Using some real engine data would be best but some quick calcs could be done to work through the temps assuming “Turbo – Intercooler – Compression” for a few different options and see what comes out as a likely MAP compensation for different CR increases?
Have you already done this?

Wingers

[bobhatton]

Winger

Do you work at Ricardo?

Bob

[Wingers]

Yep - guilty as charged...

Not too many Engineering places around Brighton I guess so it narrows it down a bit.

Wingers

[bobhatton]

I supplied nitrous parts and racing fuel to them back in the 80s and 90s
Great bunch of lads there.

Bob

[Loser]

some really interesting reading there sorry i've been quiet a few days, my mate stuffed his car into the back of a truck so we've been putting that back together, so i haven't had time to digest this and throw some calculations into the mix but i'll try to get on them tonight (22 year old on a friday night working out tdc temps, wtf is wrong with me!?)

are we saying that tdc temps are going to be the important thing here? Obviously because the engine is the same as any other 3sgte if we know what the maximum tdc temp is with one then that'll translate roughly into the maximum tdc temp for all 3sgte's, weather it be a high or low comp engine, because the way the mixture is in the cylinders and the amount it'll be heated by conduction is roughly the same?

Not been sleeping much either.. Stuff's taking a while to sink in

[Wingers]

are we saying that tdc temps are going to be the important thing here? Obviously because the engine is the same as any other 3sgte if we know what the maximum tdc temp is with one then that'll translate roughly into the maximum tdc temp for all 3sgte's, weather it be a high or low comp engine, because the way the mixture is in the cylinders and the amount it'll be heated by conduction is roughly the same?

Yep - that's my understanding of it. It’s not quite right as changing the CR will change the combustion process so it will be worse than predicted for the high CR engine but I don’t know by how much. It’s certainly a good idea to do.

some really interesting reading there sorry i've been quiet a few days, my mate stuffed his car into the back of a truck so we've been putting that back together

Hope it’s nothing more serious than some bent metal…



Wingers

[Ruishy1]

Just had a quick browse, interesting stuff. Would love to go indepth and learn all this, but tbh its soemthing i would never really use so no point

There have been various instances where high CR has been used on a charged engine ( read about it a few times, and im being lazy for not tracking down any info) and its worked, as long as the fuelling, timing and sufficient charge cooling is used you get a much better torque reading throughout the rev range,

just my 2 cents, i always like trying new things, this sounds like a plan to me mate

[Loser]

nah nothing too serious- mildly stoved in front end, hit it pretty hard though. Who needs a jig when you've got a ratchet strap and a massive gatepost
can't belive how well the bumper's come out either it's the 3rd time he's crashed it and he's never swapped the bumper, never underestimate the integrity of an mx5. Pics when i'm home

just been searching for iat readings and i've just found out you've got an iat sensor wingers, what's your setup? Sorry if this is a bit lazy but i'm supposed to be working

[Wingers]

……..never underestimate the integrity of an mx5.

Oh don’t worry I’d never do that – I used to race an mx5 so now exactly how strong they are.
Watch out for loose bonnet pins though





Anyway – what do you want to know about my setup?
I had to fit the IAT when I fitted my Link ECU. I have a rev 1 so it used to use an AFM but the Link uses MAP and IAT so I added a new sensor.

Wingers

[Loser]

Was just wondering what modifications you're running and what kind of IAT's you were seeing if you knew??

[Wingers]

plenty of mods but very little boost at the moment as I'm still in the process of sorting it out.

The key points are:
0.7bar boost
GT3076R 0.82 A/R
1000cc inectors
Link ECU
forged engine


I did take a few IAT logs a few months ago but can't remember what they were like now tbh. Generally in the 20-40 degC range iirc. The real thing to look for is the on-load temps though as they're what really matter.

I got the car back on the road yesterday though so am hoping to take a few more logs in the coming week.
Hmm, reading it back that sounds very wrong
ou get the point though....


Wingers

[Blacklightning66]

The theory behind it is that i think this is more the engine that it was supposed to be in my opinion. I read a really interesting article on the celica st165 rally tuning. Toyota team europe did a high compression low boost engine and toyota team uk chose a low compression high boost engine.

Also i'm looking at the cheapest possible way of increasing power- this method would take the ct26 closer to its efficancy island (big assumption but even at 12lbs it seems on the limit of it's depth), and i noticed that crx turbo vtecs can run 8-9 lbs safely with a compression ratio of 10.2:1, and that's on a standard engine, and they can run 12-14 psi without knock- but the engine doesn't like it.

I see what you're saying about cylinder temperatures but for the above reason i think it can be done?

Obviously if i was thinking of increasing the boost as well as the cr then yeah i agree but with me reducing the boost as well surely it could work?

Out of interest, what's the first thing to go on a rev2 after the headgasket? And when in power lol

There is a thread like this on www.mr2oc.com about this stype of setup on e85 fuel. But the CT26 and CT20b both work better at lower boost , the more boost they run, the more thye loose efficenty and surge.

I always think my 2 runs better at 10-13psi on the CT20b than when I run it at 16psi.

[Loser]

i'm gonna have to get my chargecooler set up aren't i? Run some sort of method to work out iat's and just see what i'm getting- then bump the cr to suit. Ballparking only - if i take a safe iat to be 60 degrees C with a stock cr, then work backwards from that. Seems hot but my current uprated IC definately gets hotter than that, and it hasn't blown up.. Yet

[Wingers]

I think 60C could be abit high for the standard setup tbh.

I've just checked some of my data logs and I see about 35 when driving around, increasing to 45-50 when idling. On an acceleration run it'll drop down to 25ish though.
Here's an example:
http://i210.photobucket.com/albums/bb30 ... /Run30.jpg

The i/c will get very hot and so will the intake charge at idle but as you open the throttle it should cool down a lot. I don't know how good the stock i/c is but I imagine that you'd still be suprised by the temps that it can actually output when needed.

If you're looking to get a sensor fitted to your car then try getting one of these:


It's sold by Components Ignition and the part number is XMES76. I sourced one through my local autofactors for about £12 or something like that.
I can send you the calibration points if you need as the supplier wasn't very useful when it came to providing data about the sensor. In the end I fitted a thermocouple and multimeter, then moved it from freezer to oven and back at lots of different temps.

I fitted the first one in the i/c pipe just before the throttle and this seemed to work fine in the end. You could go for fitting it into the manifold as I have since done but the difference is negligable and it's a lot harder top do so I wouldn't bother if I were you.
Here it is fitted (in the pre-throttle position):


Wingers