Knackered old Porsche with loads of natural light - Boxster!
Discussion
Heaveho said:
Interested to know where PPBB and Escy are taking the IATs from? I run a turbo car ( not a Boxster, Evo 8 modified with an Evo 9 VVT head and turbo, ecu, etc ) at 410 / 400 at 1.9 bar on high boost and measure IATs before and after the throttle body, the before using a K-type thermocouple that just hangs in the airstream where the inlet hose joins the TB, and a Lascar digital readout, the after just using the standard map sensor on the inlet manifold read through a Scangauge. I've never really been comfortable giving the car all it's got when the temps are even only in the high 30s, ( done what I can to assist it, 3" thick ETS intercooler ) so it's interesting to read how much higher yours are and that you're both ok with that. But I guess depending on where you take the readings from may show different results to mine.
There's often quite a discrepancy in the readings between the two on mine, it doesn't take long for the readings on the manifold to become noticeably higher on a closed throttle when it's obviously shut off from incoming air. I wonder how affected by heatsoak the manifold sensor becomes and how it affects the accuracy.
OE position in the inlet manifold behind the throttle body: There's often quite a discrepancy in the readings between the two on mine, it doesn't take long for the readings on the manifold to become noticeably higher on a closed throttle when it's obviously shut off from incoming air. I wonder how affected by heatsoak the manifold sensor becomes and how it affects the accuracy.
IAT sensor position can be a bit of an odd one as depending on the ECU mass of air calculation strategy it can be in a position which is less indicative of real world but is the best position for speed of response or accuracy across a wide range of air flow conditions. It doesn't matter on aftermarket ECUs as you're very unlikely to be using an OE style mass of air calculation and it's less important if using a mass air sensor as you already have a measure of the actual air mass etc. etc.
Heat soak of the sensor at low flow rates, idle engine etc. is likely an engineered element as the engine is, after all, inducting that hotter air mass so it is important for mass of air to know the temperature of the inducted air mass. In OEM land you're trying to achieve an extremely tight lambda window to keep the cats happy etc. so you're actively accounting for stuff that just doesn't matter if you're on early Euro 3 emissions or aftermarket.
The most important thing is the outcome of higher IAT's, for example are you worried about them because the engine is super close to the knock threshold and you don't have any knock control to actively manage this? Any chargecooling system that is within 20 degrees of ambient at full boost and a reasonable vehicle speed is pretty good, but engines are quite happy inducting quite high temperature air mass - you just make less power on it..... or if you're lacking all the controls to manage it end up the wrong side of the knock threshold with unhappy ring lands. Realistically it's really hard to do brilliant charge cooling in a mid engined application whereas in a front engine application (if you have no need to meet crash etc.) you generally have room to fit an enourmous air/air heat exchanger with loads of thermal mass so in mid engined installations you are always running with some amount of compromise around inlet air temps or sustainable boost.
On my car it's a bit of a double whammy of fails because I wanted whooshy turbo noises in the cabin so the air filter is hanging off the front of the turbo against the bulkhead. It does have some amount of ambient air feed but it's still a solid 15 - 20 degrees up on ambient before it even gets to the turbo, which then means the turbo is working harder to turn that already lower density air into high density air, which means it's hotter before it gets to the charge cooler, which is tiny to fit in the bay etc. etc. etc.
As I think I mentioned earlier in the thread, perfection is the enemy of good enough, I'll decide on which areas are getting developed once we have it all running happily at sensible power (400bhp'ish) and optimise from there - the important thing is that there is enough capability in the engine control to manage the sub-optimal inlet temps i.e. it has dual sensor knock control, good strategies for ignition advance management, sensible compensations etc. so I can be confident it won't go pop through knock at least. The TL;DR of all that is don't get hung up on what the IAT sensor reports, but understand what the relationship of those readings is to where the knock threshold of the engine is and how the engine control is managing that...... unless you are purely about all the power in which case worry about it very much, fit WI and send it to the moon
Thanks to both for your replies. Frankly I don't really trust the standard knock sensor to reign things in when necessary and monitor it independently via a very old school method. It's crude, but does the job. I had to turn the sensitivity down when I put the Mivec head on, as I suspect the added activity in the head was misleading it. The map's been checked professionally a couple of times recently and the knock count is minimal.
I've generally always deferred to the original map sensor reading when observing the IAT, as you say, that's the air temp entering the cylinders, and therefore, in theory, if accurate, more critical. Again, thanks to both for your insight. Great thread for lots of reasons.
I've generally always deferred to the original map sensor reading when observing the IAT, as you say, that's the air temp entering the cylinders, and therefore, in theory, if accurate, more critical. Again, thanks to both for your insight. Great thread for lots of reasons.
Decided to have a quick sniff around the back of the engine with the borescope to see if I could find my coolant weep, spotted where the gearbox was wet on top and a likely looking silicone above it and got two turns on the hoseclip holding it. If it's solved that easily then I'm off to buy a lottery ticket as it's avoided dropping the engine down to access it!
Chargecooler rad design is over with PWR who are busy making that, they've offered to do a Booster charge cooler kit as a part number so if anyone else should be crazy enough to build one at least you can order that fairly off the shelf and it'll just bolt on.
Very excited to get it on the dyno now and see what it makes, as always dyno time goes one of two ways - either smiles all round or crushing disappointment so I'm hoping for the former!
Chargecooler rad design is over with PWR who are busy making that, they've offered to do a Booster charge cooler kit as a part number so if anyone else should be crazy enough to build one at least you can order that fairly off the shelf and it'll just bolt on.
Very excited to get it on the dyno now and see what it makes, as always dyno time goes one of two ways - either smiles all round or crushing disappointment
so I'm hoping for the former! If everything works and it doesn't vomit a driveshaft out then rather boringly it will make about 400bhp @ 7000rpm and 300 ftlbs everywhere over the boost threshold..... as that's what we're calibrating to It's on standard cams so will be running a lot of boost higher up to hold the torque figure up as it will be over the top of the cams by 6K but it should make that power without too much trouble as it's well within the limits of the turbo.
I'll run at that for a bit and get everything about right in terms of systems, charge cooling etc. and then decide what the next development steps are..... I could fit an NA inlet cam, a G25-660 on a suitable manifold and some water/meth and see how long the gearbox lasts or look at a G25-550 and see how much lower we can make the boost threshold. Sensible cams for the power would probably help charge temps too as it'll need less boost to make the power, I suspect on the boost button it's going to run out of charge cooling really quickly so maybe just some water/meth and 500 on the button will do the trick.
Either way keeping it as reliable as possible with good driveability is the name of the game on this one really as 400 beans in a 1200KG Boxster is going to be pretty nippy regardless!
It's on standard cams so will be running a lot of boost higher up to hold the torque figure up as it will be over the top of the cams by 6K but it should make that power without too much trouble as it's well within the limits of the turbo. I'll run at that for a bit and get everything about right in terms of systems, charge cooling etc. and then decide what the next development steps are..... I could fit an NA inlet cam, a G25-660 on a suitable manifold and some water/meth and see how long the gearbox lasts
or look at a G25-550 and see how much lower we can make the boost threshold. Sensible cams for the power would probably help charge temps too as it'll need less boost to make the power, I suspect on the boost button it's going to run out of charge cooling really quickly so maybe just some water/meth and 500 on the button will do the trick.Either way keeping it as reliable as possible with good driveability is the name of the game on this one really as 400 beans in a 1200KG Boxster is going to be pretty nippy regardless!
poppopbangbang said:
If everything works and it doesn't vomit a driveshaft out then rather boringly it will make about 400bhp @ 7000rpm and 300 ftlbs everywhere over the boost threshold..... as that's what we're calibrating to It's on standard cams so will be running a lot of boost higher up to hold the torque figure up as it will be over the top of the cams by 6K but it should make that power without too much trouble as it's well within the limits of the turbo.
...
What kind of cost do ECUs start from that can handle this kind of arrangement? I find it really interesting how the torque can be managed into the curve shape you want and how boost can be added as the revs go beyond the cams. It's on standard cams so will be running a lot of boost higher up to hold the torque figure up as it will be over the top of the cams by 6K but it should make that power without too much trouble as it's well within the limits of the turbo. ...
I'm thinking of how a big turbo and good ECU could improve something like a 1960s fuel injected Benz (strong engine, low compression, FI inlet manifold) ....
braddo said:
What kind of cost do ECUs start from that can handle this kind of arrangement? I find it really interesting how the torque can be managed into the curve shape you want and how boost can be added as the revs go beyond the cams.
I'm thinking of how a big turbo and good ECU could improve something like a 1960s fuel injected Benz (strong engine, low compression, FI inlet manifold) ....
Really most things now have workable boost control and a good 50% support DBW throttle so you can do this sort of thing with a wide range of ECU's at varying cost points. I'm thinking of how a big turbo and good ECU could improve something like a 1960s fuel injected Benz (strong engine, low compression, FI inlet manifold) ....
The strategies that the ECU runs becomes more important the finer the control you want and to a point driveability and performance is directly related to the complexity of these strategies. You can run an engine on two sensors (crank position and load) but it won't be very nice when compared to something modern and it also won't cold start very well. On some of the more serious engine control platforms you're paying more for access to the strategies than you are for the hardware.
Torque is essentially just air mass i.e. the more air mass you have, the larger the combustion event can be in terms of gas generation and the greater the useful work on the piston crown i.e. there's more torque on the end of the crank shaft. Two things impact the air mass availability to the cylinder, one is the throttle and the other is boost pressure so with a combination of those two things you can directly impact what the torque curve looks like. This is the reason why if you plot boost pressure on a turbo charged engine the torque curve looks a very similar shape to the boost pressure plot.
Boost control can be achieved in several ways, modern production engines use electrically actuated wastegates where they can be driven to an exact position for bypass but there's nothing wrong with a decent mechanical gate. Pressure bleed off or bleed on is used via a PWM driven valve to allow boost pressures greater than the mechanical gate is sprung for. Most aftermarket ECUs will support PWM based boost control via a single solenoid valve.
If you're looking to do this on a 1960's engine though then the key challenge will be getting accurate and reliable crank position sensing in there and ideally some sort of phase sensor too. If you've got those then you can run the engine on something many decades more modern than the engine itself is.
ATM said:
Harfi88 said:
Brilliant thread, unsurprisingly the workmanship and quality is off the chart. I’ve got a hard top for my 987, I think it’s been on for two weeks out of 5 years of ownership, I think they look great on the car but I couldn’t get mine to fit quite right so it let water in on the front drivers side, I don’t remember it creaking much and made the car feel like a cayman from the inside.
poppopbangbang said:
For some reason the pic in this didn't load earlier!
Have you picked it up now?
Indeed I have. Drove home the long way and tucked it up under the rangie. I refuse to give Rishi money I don’t have to so it’s sorn till next Wednesday. Have you picked it up now?
Edited by eltax91 on Friday 26th April 07:17
eltax91 said:
poppopbangbang said:
For some reason the pic in this didn't load earlier!
Have you picked it up now?
Indeed I have. Drove home the long way and tucked it up under the rangie. I refuse to give Rishi money I don’t have to so it’s sorn till next Wednesday. Have you picked it up now?
Edited by eltax91 on Friday 26th April 07:17
wc98 said:
eltax91 said:
This one is actually in rude health and is also a dual row car, so right now no plans to do anything other than enjoy it as Porsche intended and see what, if anything, needs improving.
Ah, ok, you never know what direction these threads will go, hence me asking Gassing Station | Readers' Cars | Top of Page | What's New | My Stuff