Engine Designer Needs Tweaking

[autofrank]

I love to fool around with the engine designer. However, am I the only one who cannot reproduce any production engine with any accuracy? Torque values are always too low in the rev range, and HP is nearly impossible to duplicate without higher revving, or larger displacement, or ridiculous AF ratios.

In addition, since the update for turbos, everything seems broken. Engine parts fail like crazy at moderate RPMs even with quality at 10, (which isn't always an option under scenarios.)

I understand that producing exact copies of real-world engines is not going to happen, but it'd be nice to be close. I think the old ED (before turbo update) was closer, and it was easier to use in that parts were more tolerant of speed and displacement.

[Killrob]

Please post actual examples of engines you want to recreate and your best try to replicate them, along with the engine file.

[USDMFTW]

Please post actual examples of engines you want to recreate and your best try to replicate them, along with the engine file.

If i remember correctly, i couldent get an LS7 to spec, that said, i will try again now.

7.0L XP 16V OHVRev0.lua

(43.7 KiB) Downloaded 300 times

Quite close tbh.

[autofrank]

*delete*

[autofrank]

F23A1

This engine was used in the 1998-2002 Honda Accord LX, EX, and SE, LEV models, and in the Acura 2.3CL in North America.
Specifications

Bore × Stroke: 86.0 × 97.0 mm
Displacement: 2254 cc
Valve Configuration: SOHC, 16 valves, VTEC
Compression ratio: 9.3:1
Max power: 150 hp (112 kW) @ 5700 rpm
Max torque: 152 lb·ft (205 N m) @ 4900 rpm

[autofrank]

1ZZ-FE (Corolla - higher output: Celica and MR2)

Bore x Stroke: 79.0 x 81.5 mm
Displacement: 1.8 L (1794 cc, 225 lb)
Valve Configuration: DOHC, 16 valves, VVT
Compression: 10.0:1
Max power: 120 hp (89 kW) @ 5600 rpm - 140hp (104kW) @ 6400 rpm
Max torque: 122 ft·lb (165 N·m) of torque @ 4400 rpm - 125.8 ft·lb (170.6 N·m) of torque @ 4200 rpm

*my example is for Corolla

[Killrob]

I've taken a look at the two examples you've provided and can see what you mean, but don't understand how you can consider this as "cannot reproduce any production engine with any accuracy".

For the Corolla for instance, you peak torque earlier in the rev range, but that peak is so flat that it is basically the same value up to 4400 RPM. Also, you used forged internals, which was not necessary in this case: "cast, cast, hypercast" did the job with a quality setting of +6. It is possible to get the exact power@5600 and torque@Peak values, but indeed you need to put some fuel in. I don't see how this is a problem though, why wouldn't you put lots of fuel into the engine when running at full throttle at high RPM?

Recently I've worked out a possible additional effect/feature for the quality system which would affect the power curve via top end quality settings, basically determining the drop-off point - like when you get valve float. That is the only bugbear I still have with the engine designer - power curves not dropping steep enough when they drop. That should be fixed by that. For the rest of the NA stuff in the engine designer we are quite happy with its accuracy. Sure, there will probably be minor balance tweaks along the way, but it won't get much better without an enormous amount of effort. Getting to 95% accuracy takes as much time as getting from 95% to 98% - i.e. sadly not feasible.

[autofrank]

I've taken a look at the two examples you've provided and can see what you mean, but don't understand how you can consider this as "cannot reproduce any production engine with any accuracy".

For the Corolla for instance, you peak torque earlier in the rev range, but that peak is so flat that it is basically the same value up to 4400 RPM. Also, you used forged internals, which was not necessary in this case: "cast, cast, hypercast" did the job with a quality setting of +6. It is possible to get the exact power@5600 and torque@Peak values, but indeed you need to put some fuel in. I don't see how this is a problem though, why wouldn't you put lots of fuel into the engine when running at full throttle at high RPM?

Recently I've worked out a possible additional effect/feature for the quality system which would affect the power curve via top end quality settings, basically determining the drop-off point - like when you get valve float. That is the only bugbear I still have with the engine designer - power curves not dropping steep enough when they drop. That should be fixed by that. For the rest of the NA stuff in the engine designer we are quite happy with its accuracy. Sure, there will probably be minor balance tweaks along the way, but it won't get much better without an enormous amount of effort. Getting to 95% accuracy takes as much time as getting from 95% to 98% - i.e. sadly not feasible.

I understand. It wouldn't bother me if I wasn't reproducing real engines, which I normally don't anyway. The sim makes engines better than reality in some cases. My basic concerns are:

Torque peaks too low* for some engines. *RPM
HP peaks too high* in some engines, and difficult to change peak to lower RPM via ignition timing. *RPM
Long stroke x small bore should be more torquey vs. large bore x small stroke or "square" engine (of same capacity.) - changing these values within the same capacity does little to the hp vs. torque output.
Building engines in earlier years (40s-60s) is far more difficult now, because parts do not last even with quality 15.

Despite this, I love the concept and fooling around with the engine builder can be fun.

[Killrob]

1) Torque peaks too low* for some engines. *RPM
2) HP peaks too high* in some engines, and difficult to change peak to lower RPM via ignition timing. *RPM
3) Long stroke x small bore should be more torquey vs. large bore x small stroke or "square" engine (of same capacity.) - changing these values within the same capacity does little to the hp vs. torque output.
4) Building engines in earlier years (40s-60s) is far more difficult now, because parts do not last even with quality 15.

Despite this, I love the concept and fooling around with the engine builder can be fun.

1) Yes, this will be addressed to some extent with the top-end quality system.
2) Is connected to 1)
3) This is a myth as far as I can tell, so I think you're wrong there. It should be the other way around: capacity is directly proportional to maximum torque in first approximation. To second order approximation, you get more torque from over-square engines taking valve area into account.
4) Yes, this might still need some minor tweaking, but it needs to be difficult to build high-revving engines in that time period!

[Kubboz]

The myth of undersquare engine -> more torque stems from the fact that many engines that need to be torquey do also need to be compact and cheap, and that's what the undersquare engines are good at.

[autofrank]

The myth of undersquare engine -> more torque stems from the fact that many engines that need to be torquey do also need to be compact and cheap, and that's what the undersquare engines are good at.

It's possibly also that an engine can have larger capacity in a smaller block from increasing stroke. Either way, this is what engine manufacturers claim, and the 1zz-fe is an example. It has a 'cousin' which is nearly square and has different output.

Btw: would be nice to have a feature to delete multiple engines from the engine builder rather than one-by-one.

[Daffyflyer]

It's possibly also that an engine can have larger capacity in a smaller block from increasing stroke. Either way, this is what engine manufacturers claim, and the 1zz-fe is an example. It has a 'cousin' which is nearly square and has different output.

Exactly, that's why increasing stroke doesn't increase engine size in Automation (that is until you need to increase the bore/block size to get more stroke)

Agreed on the engine manager list, it needs a few things done to it still, like sorting, better deleting etc.