my first lp went out very early in my trucks history and the only symptom i had was a slight "stumble"on acceleration, and after 2 fuel filter changes "(about 30,000 miles) i checked the pressure and there was none.......that was the reason i had to manually fill the filter canister and bleed the injectors.........over 130,000 miles later my vp44 went out, but the fpcm failed......the guys at industrial injection said the mechanical parts look like new

 

I disagree. I will grant that there are hundreds and hundred of posts CLAIMING that a dead VP was caused by a lack of fuel pressure.

But no "proof" of cause has ever been established. It's one thing to say that a person had a dead VP and they also had no fuel pressure. But it's entirely another thing to say that had a dead VP because they had no (or low) fuel pressure.

There are literally thousands of trucks on the road as I type this with dead lift pumps and fully functional VPs. There are also plenty of cases like mine where a VP dies with perfect fuel pressure all along. Both cases seem to indicate that while fuel pressure is important for the VP to have, it is not all-important.

~~

The early VPs failed primarily in mechanical means (parts seize, stick, etc). Later VPs after 99 or so seem primarily to fail electrically (i.e., the dreaded p0216 fault code).

The electrical failures point primary and strongly towards vibration and thermal cyclic fatigue rather than lack of fuel pressure.




So, I'll restate my claim even more clearly. While we've all heard of lift pump failures and VP failures being noted at the same time, we've yet to see proof of cause.

What would proof of cause look like? It will be a graph showing VP44 life purely as a function of fuel pressure, with standard fuel specs (control lubricity variable), constant temperature (control that variable), constant duty cycle (control that variable), identical installation (gear lash, etc).

In the absence of such a kind of graph, all we have is speculation and anecdotal evidence, neither of which is proof.

I've recently seen data that shows how strongly gear lash affects vibration in a gear-driven injection pump. If you have a truck with gear lash on the wider end of the spectrum, that fuel pump experiences a lot more vibration than one with lash on the tight end of the range. This vibration factor could be a *huge* factor as to why some trucks seem to eat VPs and others seem to last just fine.


The ultimate in "proof" would be a transfer function that mathematically described VP44 life in terms of the many factors that influence it. We're clearly not anywhere near that.

The reality is that we think we know more than we did in 2001 or so, but even with all the FASSes and such sold, VPs still die.


At the end of the day, there are just too many factors that determine VP life for us to hang our hat on supply pressure. At the very best, it is one of only 6 or more factors that have large influence on VP operating environment and therefore component life.

My take on this has changed slightly since I started working for Cummins, talking to people who developed the old VP engine, and have done a lot of work on fuel system related components (pumps, lines, injectors, etc).

I used to think like everyone else, that the internet forum wisdom was gospel truth.

But now that I can better see the situation overall and have access to real data (I actually have the engineering drawing for the HO and SO VPs), my perspective has changed, and I'd like to think I've improved my grasp of the situation.

JMO

 

I respect your response except that you mentioned where you're not agreeing with my fuel pressure statements but coincidentally agreeing with me by then stating that there is NO data to help us understand the VP any better then we currently do. So I'm taking it that you too think that there's something specific governing the VP failure rates but that you also think it has something to do with the engine instead of the pump itself or fuel. Also, if what you say is true about commingling with individuals whom happened to be part of the VP/ISB conception, then clearly you should have more of an idea than what you've posted. Unless those said people really had nothing to do with the VP's R and D itself but rather were there simply to mate the VP to the ISB. Although, I have to say that I'm not fully disagreeing with you when you're suggesting that there could be many other variables which possibly play a part in the VP failure statistics. Sure, possibly there is factors that may aggravate others but I think not to the level you're insinuating. If that was the case then other injection pumps would have similar failure rates as the VP while bolted to the Cummins? Or did Cummins redesign anything to compromise or accommodate injection pump reliability after 1998 and after 2002? Also, why isn't there more ECM failures? Those electronics are subject to the same vibrations and heat as the FPCM.

Well nonetheless.....given your strong feelings about the lack of need for necessary fuel pressure to the VP, you had to have known that I was going to say this next.......... If you feel you're right then why dont you simply disconnect or remove your fuel pump and see how long your VP lasts? I'm not disagreeing with you that the VP will continue to run with zero fuel pressure but that would be a simple and effective test to determine how long the VP will last with zero positive fuel pressure placed on the injection pump. If your VP continued to last thousands and thousands of miles on no fuel pressure, as you claim which thousands of trucks currently on the road are running at, then you too will have "proven" that the VP does not need any fuel pressure to function properly or for that matter, function at all.

And please dont take my post as me trying to be cynical or facetious either. I truly enjoy debates like this since opposition breeds ideas and for the most part people dont like confrontation so they tend to stray from just that, but unfortunately we've quickly come to the end of cross speculation. So I feel that there is no other way to really know anymore unless some hard data is created. I'm currently running a Raptor at 19 psi idle and 16 psi WOT through 1/2" fuel lines on a VP which is about 2 years old. I will continue to run like this on everyday ULSD with no additives, and then for comparison you either disconnect your fuel pump or remove it altogether and run with little or no fuel pressure and we'll simply see who runs the longest. Maybe mine will quit.....who knows.

 

Not at all. I'm not at all saying it's engine OR it's fuel pressure. I'm saying it's all of it, and there are simply too many variables to know for sure. Fuel pressure is certainly a factor-- one of tons of factors. But so is vibration. So is thermal cycling. So it part-to-part variation. So is gear lash. Etc etc.

Since Cummins bought the pump from Bosch, it's fair to say that no Cummins employee developed the pump. But they would communicate with Bosch and they can share with me their experience about what was noted in development. Mating the VP is a good way to think of it, though Cummins does its own validation of all parts bought out finished. As for what I should or shouldn't know, that's a matter of opinion.


This last statement is completely illogical, as it assumes all injection pump are of equal reliability and are equally robust against the myriad factors that attack them. Or is this assumption part of your point (i.e., all pumps aren't identical) It seems illogical to me, but I might be misreading your intent.

Those electronics certainly are NOT subject to the same vibes. Different portions of the engine vibrate differently. Gear trains carry a lot more vibrations (cam/crank torsionals), so anything driven by them experiences vibration different than other places.
Similarly, you can take the same thing and bolt it to 20 different places on the engine and get 20 different vibration levels. Generally, closer to the crank centerline and closer to the middle of the engine means lower vibes.

I've NEVER said that fuel pressure is unnecessary. Putting words in my mouth is unfair and I don't appreciate it. Fuel pressure is a good idea- no doubt in my mind. But no one knows how much is enough, how much is too much, and whether there's any difference at all between 5psi and 15psi and 25psi in terms of VP life.

At least, no one here in internet gearhead land. Certainly someone at Bosch probably knows, but they haven't seen fit to join our little club and share.


Thanks for engaging a useful discussion.

 

I should have been more clear in this post. The absence of ANY fuel pressure isn't good. But I was meaning that the absence of FP over any particular point-- say 5psi, or 10psi. We all agree that fuel pressure is a good idea. But we have no idea how little is too little.

I'm of the mind that any fuel pressure at all is enough.

 

I would NOT use wmo. I gave it a try, and found that it "coked" up my precombustion chambers BADLY on my excavator. Yanmar IDI engine- fixed it by pulling injectors and scratching it off with a coathangar, and blowing it out while using compressed air while turning the engine over. I suspect it is zinc oxide an anti-wear addative in motor oil.

The engine was getting hot so easily, that I thought I'd blown the engine up. Probably because the air wasn't spinning in the precombustion chamber due to the massive amount of fouling. Looked like about two (!!!) tablespoons of chalk in each precombustion chamber.

Just a caution.

BTW blue chip diesel has a bit to say about vp failures, as well as how they address them. www.bluechipdiesel.com worth a read regardless of where you get your new vp.

 

Well now you say this..... Certainly changes most of my opinions on what you've been saying too. Although, I am still an advocate for what I believe the optimal fuel pressures are to be for the VP. I wont go though the whole thing again, so in short, I think that the FP needs to be above the overflow valves setting to assure a constant rate of fuel cooling the VP as that excess fuel passes through. Thats why I chose 19 idle. This way I'm not overloading the return lines with an onslaught of excess fuel volume but that there's always enough excess to transfer heat away and my WOT pressures stay at or just above the overflow valves settings. How do I know my theory is working? Well when its 110* outside, I can easily touch the top of the VP and even hold my hand on it.....all while the rest of the engine is around 200*. I think that there says a lot. Sure a thermo laser gauge would be more accurate but it still proves a definite reduction in heat. The only problem is that when the engine is shut down, I still dont have anyway to keep the engine heat from soaking the VP. And there's also other problems with the fuel module which fights against this too, which is also why its crucial to have no less than 1/2 to 1/3 tank at all times in the summer months, but lets save discussion that for later.

And I also agree with you that there's more variables which are deeming the VP as an unreliable pump. But I'm not convinced that the VP itself isn't a reliable pump. Yes, Bosch gave Cummins the pump but it was up to Cummins and Dodge to put the rest of (mate) the package together. Fuel lines, fuel pump, fuel tank, fuel module, vibrations, heat, etc, is all Dodge and Cummins doings. Which leads my to think that because the VP was only a 4 year "temporary fix", neither Bosch, Cummins, or Dodge really took the time to put together the proper package for the VP. Its not like that doesn't ever happen in the manufacturing industry. As I said before, maybe they had no idea how popular the VP would become with aftermarket enthusiasts.

Now I know you already know all this and I'm being redundant too but I still think my theory is a valid theory and should be considered and not be passed off as illogical. And after talking to a couple VP reman vendors, I understand that they're thinking along the same lines too. Plus, if fuel temps weren't an issue then Bosch wouldn't have installed a fuel temp monitor in the VP. So in point, I'm not saying that the VP "needs" that much FP to work properly. Matter of fact, I too think the VP will work just fine at much lower FP settings (ie. 5, 10, 15 psi). And the truck will run and run well too at those lower fuel pressures as long as the VP has a constant positive pressure supplied throughout all fueling demands. How low does the FP have to get before the diaphragm is compromised? Not sure, BUT.....again, thats never been my argument. The ONLY part I care about is when the FP is below the overflows setting then there's no fuel cooling the VP. The P-pump was cooled by oil but the VP is not and unfortunately by design, there's nothing left to keep the VP's temperatures from reaching the same temps as the block.....which are detrimental to the lead free solder connections in the FPCU.

In short, when I thought you felt that NO fuel pressure was OK is what spurred on my posts.

 

YOU R CORECT

Main reason for VP to go bad is poor fuel.
The tolerance in the VP pump is so small that the fuel needs to be perfect clean

If you total disconnect the lift pump and get a bottle wit fuel and connect it to the VP pump with hose and it will run with no problem

 

The excess fuel doesn't pass through. You cannot increase fuel flow through the VP with increased pressure- period. Whether you run 18psi to the pump or 30psi to the pump, you'll still get about 30gph returned back to the tank. If you cannot increase flow through the pump, it follows that you cannot increase cooling to the pump with increased pressure.

Then add the fact that higher fuel pressure means warmer fuel temps, and you further erase any cooling benefit to higher pressure-- to the point where more pressure can *increase* pump operating temps if taken to the extreme (30psi or something nuts like that).

One good explanation for the measured lack of higher flow at higher pressure is that the inlet to the VP44 has a positive displacement gear pump of sorts built in. Thus, you can't force it to flow more (fixed displacement). There may be an "overflow" of some kind built in, but it's not before the gear pump. Thus, you are not increasing "overflow" flow with higher pressure. You're not increasing the flow of ANYTHING with higher pressure. With not increase in flow, there's no increase in cooling.

That is the flaw with your theory-- it does not agree with empirically measured data in terms of return flow.

As for hands-on temp measurement, I don't think that really tells you anything. For one, how do you know that it hasn't always run that cool? How thick are your calluses? So on and so forth. You can measure things, or you can guess. Nothing wrong with either, as long as they are kept within their appropriate places.

Finally-- I just checked the engineering print for the HO VP44. It says inlet pressure tops out at 0.8Bar (that's about 11.6psi).

If you are a TDR member, the old "pumps, lines, what not" thread contained the info on not being able to increase drain flow with higher VP inlet pressure.

Or, just call Bill Kondolay at DTT and ask if he remembers the work he did with flowmeters and such --MEASURING return flow at different inlet pressures.

Justin

 

Not true Justin. Excess fuel does indeed pass through the VP. For starters, if it didn’t then what would be the reason for the overflow valve, and the 1/16" hole in the overflow valve? Why not just let the internally regulated lift pump/fuel pump feed the VP all the fuel which the engine needs and leave it at that? And, yes, you certainly can increase the flow of fuel which passes through the VP by increasing the pressure, but only to a certain point, then you reach a limit to how much volume which the VP can actually flow. And the extra fuel that does pass through absolutely does aid in cooling the pump. Now whether pumping 18 psi or 30 psi will yield you the same back to the tank, I don’t think that’s the case at all. I can’t say how much return is lost from pressure reduction, volume, engine usage, or friction, etc…... I’m not sure anyone could answer that either given the numerous variables involved and the circumstances of a specific application. Point being is that the VP is indeed a fuel cooled pump, unlike its predecessor, and there’s no way that the small amount of fuel which only the engine requires is going to be able to keep fuel and pump temperatures down. Another point is the fact that Bosch equipped the VP with a fuel temperature sensor to compensate for varying fuel temps, but the VP will also reduce engine power if those fuel temperatures get too high. I want to say that number is over 165* but I just don’t remember at the moment. For what reason would the fuel temps ever climb to what Bosch considered an unfriendly level if there is always the same amount of fuel in the VP at any given time? So if your fuel temp gets hot then simply floor it so more volume passes through the VP in order to get all that overly hot stuff out?????

Moving on….. Yes, higher fuel pressures is going to create warmer fuel…..but how much higher is the fuel temp going to be simply because you’re fuel pressure is a few pounds over OEM specifications? Or for that matter, at what pressure would the fuel temp be? Again, I’m not sure anyone could answer those questions while leaving out all other factors and variables involved but I’d wager to say that fuel temps don’t climb too much until you’re dealing with pressures far above fuel pump capabilities. Also, I never said anything where running 30 psi of FP was a good idea. I’ve only had one opinion regarding fuel pressures and that is no more than 20 psi at idle and no less than 14-16 at WOT. That’s far from 30 psi I think you’re insinuating I said or ever suggested. Sure some guys do opt to run over 20 psi but that’s their choice.

As for your explanation of how the VP inlet works……….you’re right by saying that it is a combination of sorts, but it’s also not gear pumped. It’s a rotary/distributor pump and the rotor on the shaft has a slot in it and so does the housing. When the two slots align with one another then fuel is allowed to flow into 3 separate passageways, of sorts, where the fuel pressure is increased until delivered to the cylinders. Yes its fixed displacement but not where the rotor is. The fixed displacement is in the three other passageways which are specific to metering fuel. All fuel must pass by the rotor first in order to assure adequate lubrication of the rotor, and then all the other internals are also lubricated by the fuel as well. If the fuel wasn’t allowed to flow through the entire pump then there would be components which would never get more than a tiny amount of fuel to keep them from screeching to a halt. Again, if this wasn’t true then there’s no need for the overflow valve because the overflow valve isn’t directly connected to the inlet. The fuel has to pass through the internals of the VP to reach the return outlet. If you don’t believe me then unhook your return line somewhere after the overflow valve and cycle the fuel pump. You’ll see that nothing should come out unless you start the engine. This is because when the VP rotor shaft isn’t turning and those slots aren’t aligning then no fuel is allowed to pass though the pump. So my “theory” isn’t flawed because its fact.

Then to talk about my “hands on” experiment. Sure it absolutely tells me something. It tells me that by simply placing my bare hand/fingers on the VP, I can detect that it is not as hot as the rest of the components on the engine, which happens to be at a toasty 195*-200*. But when the engine is off, the VP eventually (about 15-30 minutes later) reaches the same temp as the engine. As you may have remembered, my experiment took place while idling after driving in triple digit summer heat. If the VP wasn’t being cooled, then by all means, it would have been the same temperature as any other piece of metal under that hood……….which were far too hot to touch, especially one bolted directly to the side of the block. As I also mentioned before, sure it’s not as scientific as a laser thermometer but I guarantee that my blue collar hands aren’t so callused over that I can’t differentiate a 50* temperature.

I’d love to see the engineering print for the VP too. If you can, link it. I can’t say without actually reading what it says but I’m certain that the 0.8 Bar is referring to not the total inlet pressure allowed in the VP but more on the amount of pressure during cranking or simply the maximum required running pressure. I’ve never disagreed that the VP will allow the engine to run on almost zero fuel pressure but that doesn’t mean it’s good for diaphragm or that there isn’t going to be a serious lack of cooling to the VP. Remember that low fuel pressure causes the diaphragm to crack, which happens because on the pump side of the diaphragm the pressure is 130-135 psi and on the fuel pump side is around 10-20 psi. When the fuel pump side of the diaphragm drops to below those pressure numbers then the diaphragm vibrates violently as the VP solenoid sends pulses signals, eventually compromising diaphragm integrity.

When I get a moment, I wouldn’t mind calling Bill Kondolay to hear the exact process of the testing he was involved in. I don’t ever claim to be smarter than anyone and anywhere I can collect more data regarding the VP, the better.

As always, it was good talking with you.

 

Here is a diagram of the VP44 injection pump showing the internal components.

On the left side of the pump, the RED arrow, is where low pressure fuel from the lift pump enters the VP44. This fuel travels down to the rotary vane pump input. There is also a passage where higher pressure fuel, shown in ORANGE, gets fed back to the input above where the lift pump fuel comes in. All the fuel entering the VP44 goes through the internal vane pump.

Fuel leaving the internal lift pump, shown in ORANGE, is at a higher pressure than the lift pump and travels to the accumulator, top right, and the timing piston, bottom right.

The spillover valve returns fuel to the tank and this is located above the accumulator in this diagram. This spillover valve only gets fuel from the accumulator and high-pressure solenoid. It is this fuel that is responsible for cooling the VP44.

The high pressure fuel, shown in YELLOW, come from the three piston, high pressure rotary pump. It is this fuel that will over come the pop pressure of the injector and be injected into the cylinder.

There is no path for the fuel to enter the VP44 and make its way to the tank if the pump is not rotating.

 

Good post, Katoom. The extra content to your previous comments helps clarify what you have been saying (and helping my obtuseness).

While I have access to the top level engineering drawing for the VP, it's pretty worthless at showing operation as it doesn't have the parts blown up or PNs listed for sub level components. So it lists generic specs (fuel temp ranges, quantity of fuel delivered (mg/stk), and inlet pressure range). I suspect the inlet pressure specs on the VP are what the lift pump is expected to deliver-- there's no context on the drawing. Anyway, it's lists a max and a min-- and I've already posted what the max is. Why it's set there would be just wild speculation on my part.

Thanks for expanding on the VP theory of operation. Quite frankly, my knowledge of its internals is pretty minor. So the extra knowledge you can contribute to understanding its operation is very helpful.

If your satisfied that extra fuel pressure is cooling your VP, I won't attempt to dissuade you.

The max acceptable inlet temp (on the print) for the VP is 71C (about 160F). A fuel temp sensor in the VP makes sense because the change in viscosity affects the leakage of the VP pistons, and thus, the quantity of fuel injected is likely reduced at elevated fuel temps. Some means of compensation would seem appropriate to me.

No ones implying or saying you run 30psi. It was only used as a reference to where fuel temps can be toasty from excessive pressure. No need to misconstrue.

Let me summarize my argument in more concise form:

I claim your theory of increased cooling flow with increased delivery pressure is false exclusively based on the DTT testing Bill K did back in the day (i.e., the famous "pumps, lines, what not" thread). I *vividly* remember that they could NOT increased measured fuel return with fuel pressures well over 20psi.

So I put it to you-- you have a theory that increased delivery pressure increases fuel cooling.

I have a fact that increased delivery pressure did not result in a greater quantity of fuel, as measured empirically with some rather pricey flowmeters by Bill K. From this fact, I extrapolate that no additional cooling can be obtained with higher inlet pressures because flow is the cooling medium-- more flow, more cooling; less flow, less cooling.

Now I suppose it may be possible for the increased delivery pressure to improve VP cooling without producing an increase in return flow. But I am unable to conceive a means by which this is possible.

It's rather clear now that my understanding of the VPs operation is pretty bad. But one need not know ANYTHING about the VPs internal operation to solve a basic thermodynamic equation of heat energy and mass flow in vs out of the fuel pump.

I appreciate that anecdotal observations carry a lot of weight. I rely upon them myself all the time. Given the same anecdotal observations on my truck, I'm confident I'd feel similar to the way you do. But anecdotes are not measurements, and as evidence goes, they are pretty weak.

I've said all I can to argue my case. Again, the key points I make:
-- increased VP cooling can only occur of extra pressure results in more drain flow
-- I vividly remember empirical testing by Bill K that demonstrated that increased drain flow from the VP does *not* occur with increased inlet pressures.

I could be remembering the thread incorrectly, but it's VERY clear to me, and I'm 100% confident. If I'm wrong, I'd be 100% wrong-- and more than a little embarrassed.

Let me see if I an get the 'way back machine' to cough up that amazing TDR thread.

 

Here's a post from me when my VP died, referring to the "pumps lines what not" thread. So I was similarly convinced (or wrong) back in 2005..

https://www.dieseltruckresource.com/...ap-t70661.html

 

I have a PM in to Stephen Kondolay on another forum, maybe he will have the access to the original VP44 return flow test data from the "pumps, lines, what not" thread.

Google only gives you hits of me talking about that thread, not the thread itself!

oy.