I won't repeat them all, but I'll try to address most of your points. This reference gives some of the details used in my discussion below...

Let's Talk Injection Timing, TQ, HP, & EGT...
https://www.ford-trucks.com/forums/7...p-and-egt.html

...and as shown in the pic in post #1, there's a critical time interval between the start of fuel injection, the start of combustion, and the build up to peak cylinder pressure. This peak pressure should occur at about 20*ATDC for any RPM and load condition, and the injection timing and pulse duration needs to be adjusted to the proper values to accomplish this.

A gasser ingests a fairly homogenous mixture of fuel and air into multiple cylinders at the same time so no matter how high the RPM there's always a combustible mixture ready to fire when the time is right, but a gasser only fires one cylinder at a time, and the timing advance depends on RPM and load (ignition delay depends on the amount of fuel) so as to maintain a peak cylinder pressure at about 20* ATDC.

A diesel ingests only air into multiple cylinders at the same time, and then must inject the fuel so that it auto ignites and then reaches peak cylinder pressure at the correct time. Auto-ignition requires some chemical pre-reactions which in turn require some heat due to compressing the air.

If fuel is injected too early, there's not enough heat of compression developed to vaporize the fuel and you get a lot of cylinder wash, and peak pressure rises too soon and places a large force on the rod while it's near TDC. If the PW is too long so that fuel is injected too late, it's still burning during the exhaust stroke which leads to excessive EGT. There's only a limited interval of crankshaft degrees over which fuel can be safely injected, which I've estimated to be about 10*.

At 3000 RPM 1*=55.55 us of time delay, so a typical ignition delay of 1000 us is 18*, and an additional 500 us of delay to achieve peak pressure is 9*, and assuming the injection pulse is about 555 us (10*) in duration, an injection advance of about 18+9+10-20 = 17* BTDC should give the peak pressure at about 20* ATDC, and this in turn should optimize the MEP at this RPM.

In the above example fuel is injected from 17* BTDC to 7* BTDC, and if you include an additional 560 us time delay for a stock IDM to turn on a stock injector, you see that a total IDM output pulse of only 1115 us is all that's required to inject fuel over this 10* interval.

At your example of 3700 RPM, 1*=45 us so that a PW=450 us is all that's required if fuel is injected over a 10* interval, and the additional 560 us time delay for a stock IDM gives 1010 us. Now at 1500 RPM, this becomes 2 x 555 us + 560 us = 1670 us as the required IDM output pulse duration, which is why I claimed at higher RPM you want a shorter pulse duration, not a longer one.

The RPM limitation of a diesel is due to the relatively long time delay that's required for auto-ignition and the build up to peak pressure, which is typically about 1000 us and 500 us respectively. At lower RPM like 1500, this requires only an additional 13.5* advance, but at 6000 RPM there's a 2x18 = 36* delay due to auto-ignition, and a 2x9 = 18* delay to reach peak pressure! At 3000 RPM, you can start injection at 17* BTDC and make peak pressure at 20* ATDC, but at 6000 RPM you'd have to start injection at about 36+18+10-20 = 44* BTDC, and there's not enough heat of compression developed at 44* BTDC to avoid cylinder wall wash down.

This is one reason why multiple injections per combustion event are used in newer engines, so that small squirts of fuel can be injected early on without washing down the cylinder wall. Also, modern CR injection uses 30,000 psi so they can inject more fuel in the shorter time that's available at higher RPM.

The newest approach for diesels is the HCCI= Homogeneous Charge Compression Ignition engine which is somewhat like a gasser in that it uses a premixed homogeneous fuel-air charge that's then compressed to cause auto-ignition, but unlike the heterogeneous mixture which results from current fuel injectors, the homogeneous fuel-air charge ignites very quickly somewhat like in a spark ignition gasser.

Now, an 8 cylinder 4 cycle engine has one power stroke every 90*, and fuel can only safely be injected over about a 10* interval during each power stroke. So, just how do you propose to fire more than one injector at a time, and could you present a candidate timing for each injector firing in both crankshaft degrees and IDM pulse durations like I did in my examples.

 

cool man and while I was watcing that I found this one that cracked me up.

YouTube - 2002 Ford F-250 Powerstroke Diesel Pulls Semi

 

(don't argue with this man), I give up.

 

What he said.

Seriously, I appreciate everything said here, and agree 100%, but I think part of the fun of this thread was the simplicity and the ability to actually see and control the injectors like little oil splitting slaves.

 

I've got no problem with trying to push as much air and fuel into an engine as possible to see how much HP it will make, and yes, the ability to run at higher RPM can definitely increase HP. However, the engineer in me suggests that the best approach is to first try and understand what the fundamental limitations are. I learned this the hard way trying to squeeze more HP from my little sports car engines.

At 3700 RPM, 1*=45 us, and since an 8 cylinder engine fires once every 90* that means the IDM needs to generate an output pulse every 4 ms, however to inject fuel over a 10* interval at 3700 RPM the actual fuel injection event is only 450 us long. Another way to say this is that the IDM needs to generate 250 pulses per second to maintain 3700 RPM, but each of these pulses only injects fuel over a 450 us interval.

Each injector only cycles once every 720*, which at 3700 RPM is once every 32 ms, or 31 times per second. My research into the IDM mod indicated that injectors turn on in 560 us with a stock IDM, and if you do the mod they turn on in 440 us, so comparing that to 32 ms, I don't see how the injector cycle rate is a limitation to achieving higher RPM. Even at 7400 RPM, the injectors only need to cycle once every 16 ms, or 62 times per second.

Now, if the stock IDM can't generate more than 250 pulses per second that definitely limits the RPM to 3700, but if you mod the IDM to overcome this limit, you also need to insure that fuel isn't injected at an inappropriate time or you'll definitely hole the block! You want to inject the fuel so that you achieve peak cylinder pressure between about 15* to 20* ATDC, where 20* is a more conservative value. You can also push the 10* injection interval some, but I've already mentioned why injecting too early or too late is hard on the engine.

The timing accuracy of that crappy CPS is an additional limitation to achieving higher RPM on a 7.3L PSD! Since 1*=45 us at 3700 RPM, you need a +/- 22.5 us timing accuracy from the CPS to inject fuel with a 1* accuracy at 3700 RPM. Also, if the nominal turn on time delay of 560 us varies say by just 10% due to temperature, etc..., that amounts to a 1.2* timing error at 3700 RPM, and a 1.6* error at 5000 RPM.

With only one intake valve and those crappy stock plenums, I doubt that you can force enough air into the cylinders to move the peak HP up to even 3700 RPM, much less higher than that. For all of the above reasons, I suggest that if 3700 RPM is a hard limit of the IDM, to leave well enough alone, and that trying to push past this limit is probably a good way to hole the block, especially with PMR's. Remember, it's not HP per se that breaks rods, it's the force that's applied to them, and if this force isn't applied in the vicinity of 15* to 20* ATDC it isn't producing max HP it's only needlessly breaking your rods!