Based on my "listening" experience with my old F350 using a DP 40/60 TOW setting and hearing 1-2 seconds of "injector clatter" after towing a grade and letting off the throttle quickly and on my digital read out "histogram" measurements of HPOP ICP and LPOP pressure on my CAT C7 when doing the same thing I think that those with AE should be able to come up with a "standard road test" for identifying bad injectors that are leaking too much oil.

The graph below is an example of what I see with my C7 if I transition from towing on the flat to pulling a grade and then let off the throttle quickly near the top of the hill so as to quickly return the RPM to idle. If I ease off the throttle and slowly return the RPM to idle I don't see an overshoot in ICP followed by a bleed down time but rather a smooth decrease in ICP from the initial 2,700 psi at 1,600 RPM to 1,500 psi at 700 RPM.

For some reason the graph won't display so you'll have to click on the link and see it in a separate window!
[img]http://h1.ripway.com/ernesteugene/DG5/Injector.jpg [/img]

It seems to me that once the oil rail is fully pressurized the only way the ICP can bleed down to the new lower value being commanded by the ECM at idle is through the injectors and that if you've got leaky injectors this overshoot will be less pronounced and the bleed down time constant Tb will be shorter than for good injectors and maybe for very leaky injectors the overshoot and bleed down time constant Tb will be will be zero just like letting off the throttle slowly allows the ICP to bleed down to the lower value without an overshoot.

As shown in the lower curve it's also clear that the VOF=Volume Oil Flow gal/min=gpm from my C7 LPOP can't quite meet the maximum VOF demand of the HPOP. Since my LPOP pressure is the same as it was on my factory engine dyno sheet I don't think I've got a weak LPOP with only 22K miles on my engine.

The above seems to suggest that the Ford LPOP probably doesn't have a lot of extra reserve VOF capability to meet the increased VOF demand of upgraded HPOPs and larger injectors? The LPOP and HPOP are both constant volume pumps where the VOF is proportional to RPM but at higher RPM and WOT the injectors demand disproportionally more VOF than they do at lower RPM.

 

If I post it here it displays???

 

I wonder how easy this would be for us since many see different peak or sustained ICP readings. Sounds like you would almost have to have a known good reading from the truck in question to compare to for future reference, unless the peak and drop down is consistent across different pressure levels on the graph.

I'm not sure we have any LPOP issues, since several have run big oil with now supply problems. I would think that since the increase in HPO is confined to a relatively small percentage of overall oil and space, LPOP volume should be fine.

Any guesses on how the graph would be affected by running empty as opposed to pulling the trailer? Got any test ideas for those of us who don't have any hills?

 

As indicated below an ICP vs Time bleed down curve is like the decreasing Voltage curve on a Resistor Capacitor circuit which has a "unique" RC time constant that depends only on the value of the Resistor and the Capacitor.

When the throttle is returned to idle the ECM shuts off the injectors so that the only way the ICP due to the residual high pressure oil that's stored in the oil rail and feed lines can bleed down is through the leak rate provided by the injectors.

The Capacitor is the volume of the oil rail and feed lines and the Resistor is the injector leak rate. If you use AE to data log some ICP vs Time bleed down curves with the throttle at idle and then look at the 36.8% Time you get a "unique" bleed down Time Constant Tb that can be monitored for a given truck and compared from one truck to another. This Tb should be the same for all identical HPOP and injector configurations!

The rate of decrease in ICP is fractional in that it takes a Time=Tb for the ICP to decrease by a factor of 0.368 from its previous value so that you can look at the ICP at any value say ICP1 at any point on the curve and note that Time T1 and then measure the Time T2 for the ICP to hit a value of ICP2=(0.368)(ICP1) and get the Tb=(T2-T1).

 

I never considered a leak rate in the injectors. I always assumed pressure would bleed backwards through the IPR since that's what controls the pressure in the oil rails. Seems to me like if the injectors leaked some in the beginning by design, the leak would consistently get worse due to the pressures we're dealing with here.

Thanks for the explanation of the graph fall off. Starting numbers may be higher or lower, but the time and rate of fall (& spike) is what we're looking for.

 

Yes I just found a diagram of the 7.3L HPOP and on the high pressure side there's a drain passage that's modulated by a piston in the IPR. I'm wondering if my C7 has a similar feature because I can't find info on it.

When the ECM commands a maximum ICP the IPR piston completely covers the drain passage and I'm wondering just how quickly that piston can react and uncover the passage to relieve the pressure? I definitely see a spike in ICP and a bleed down when I snap the throttle back to idle!

 

I was thinking the same thing. Can you elaborate in this area Gene? If there is a bleed down, what happens if they clog or sludge up? Do you then start to blow out o rings?

Edit, well never mind then. I see you were typing while I was reading the thread.

 

I think your original assessment with the graph still works, it's just a different bleed down point. The question now is, does our IPR react faster than whatever drain you have.

I would picture a leaking o-ring or injector removing the spike, and having a steeper drop off since the bleed off would be instantaneous, instead of reactionary.

 

Yesterday we towed from Laughlin NV to Kingman AZ going the long route down to I-40 and crossing the Colorado River then gaining 4K ft in altitude. On the way up I played with the ICP overshoot some but without releasing the throttle to idle as I've done before when slowing to enter a rest stop or a brake check area at the top of a grade.

I watched the TPS and the ICPS on the same display and when I backed back off the throttle from 90% TPS to 40% TPS in less than 0.5 sec I saw an overshoot in ICP followed by about a 1 sec bleed down to a new steady state ICP value. If I eased off the throttle from 90% TPS to 40% TPS in about 1 sec or longer I didn't see an overshoot in ICP. When I'm towing I don't need a grade to see this effect as I can just let off the throttle from a given cruising speed but the overshoot in ICP is more pronounced if you're at a nearly WOT steady state condition and then let off the throttle quickly.

The reason I did this post was because I read the threads on HOP pressure measurements and people were wondering if the lower than expected ICP readings with their new HPOP might be due to a bad injector "pissing" too much oil and I thought that looking for an overshoot in ICP might be a way to find out without pulling the valve covers.

If it turns out that all I'm seeing is the reaction time of the IPR that also might be worth investigating and understanding better to develop a "standard" road test for that.

Note to self... the last week in March is too early to be heading eastbound on I-40!

 

I can only imagine what the guy behind you must have been thinking as you kept taking it up for short WOT runs only to let off the accelerator.

We do need to do some testing on this to see how the PSD reacts to this. My biggest concern is without the additional load on the engine from pulling a trailer, the spike is likely to be small, quick, or possibly non-existent in my mind.

But as long as we've got all these new pump guys out running a bunch of WOT runs to compare before and after pressure, it sure would be easy to just let up off the skinny pedal while the data log is running. Then we could have Barney do the same and find out if there's anything to be seen from this test on a 7.3.