Hello all,

I tried posting some days back on the 'diesel in distress' section but haven't heard a peep so I thought I'd try the 2nd Gen Dodge Cummins section . In the interim since I initially posed my inquiry, I ordered and purchased a TorkTek overflow valve for stock applications assuming that maybe the overflow valve was bad and bleeding all the lift pump pressure off, hence the lack of steady pressure but to no avail. It turns out the truck already had an adjustable overflow valve of some sort installed (no markings) but a new valve did not stop the gauge needle bounce problem. When getting the valve and install instructions, their comments about needing a snubber if a person wanted to add a temp or permanent gauge to monitor lift pump pressure is the first mention I've found ANYWHERE about excessive needle bounce on a gauge put on the lift pump line to the injector pump, but still no answer to my Q and TorkTex has no way to contact them by email or ph, and my cc statement listed a ph # from Romania as a credit card contact # and leaving voicemails haven't had any luck.

Anyway, I did get a bit better pressure with the new pre-set valve that they say is set at 30psi, but my gauge needle still bounces from about 0 to now about 50-60psi with each lift pump piston cycle. Is this 'normal', or is something still amiss? can an injection pump run safely with such an extreeme pulsation of pressure to it? Should I, could I add an inline check valve of sort? It almost acts like the lift pump has no check valve function and as soon as the piston pump cycle pressure is over, the pressure falls back to 0. Older gas engine mechanical fuel pumps used to act this way if the internal check valves would begin failing but I am not that intimately familiar with diesel injection pump functionalities on the 12 valve cummins. Following is my original post earlier.

I did an extensive mod so I could run waste oil on a 96 Dodge 250 5.9 with about 240,000 mi on it a few years ago, but have been running diesel due to current low cost. In all my searches online I see a lot of people with lift pump pressure problems due to the overflow valve but no comments about my problem.

I have a pressure gauge on the line just before the injection pump. When I completed my mods (see below), I was still getting about 30psi with mild needle wiggle that was in sync with each engine revolution/lift pump pulse. I ran maybe 1/2 tank of 90/10 ratio WMO/diesel filtered down to one micron just to test to make sure the mod worked well, and actually got better power and fuel economy with the WMO but stayed with diesel due to fuel prices being down and until I can find someone willing to part with their WMO other than what little I produce on my own.

Over time the needle wiggle got to be more of a bounce, then worse to where needle would bounce from 0 to 30 with each pulse, then to where it would really bang the needle on 0 and barely lift up to maybe 5psi on the bounce. Concerns of not wanting to damage the injection pump but not knowing for sure where the problem lie, I replaced both inline filters (an add-on pre-heater/filter unit that is just after the tank as well as the stock filter) and also replaced the lift pump. I got better pressure and the needle bounce goes from about 0-5psi to up to about 40psi but the needle bounce is still very strong. The degradation and the needle bounce issue occurred over probably 3-5000 mi.

Has my overflow valve worn out/gone bad to where it's bleeding off most all the pressure the lift pump produces creating the excessive needle bounce? (hence replacing the overflow valve now, indicates that isn't where the problem is)...

Following is mods I've done and a few stats. Oil pressure at idle is 30psi and about 60psi @ 1,700rpm with engine warm. The kid that had it before put on an adjustable pot to control the overdrive engagement and someone told me that the strong engine shudder that occurs with gentle acceleration from stop may be due to him messing with the throttle plate as a means to increase HP (not sure).

I took out the stock fuel heater due to it's inherent leaking problem. Fuel flow from tank goes first into a custom electric thermostat controlled heater block that kicks out once the water heater filter/warmer warms the fuel up a bit that is a combination hot water heater/filter unit that filters down to probably 7-8 microns and heats the fuel all the time. Then it goes into the lift pump and then into the stock filter. From there it goes into a second electric thermostat controlled heater block that is just before the injector that kicks in only if fuel / engine is cold. Keep in mind that initially after my mods, I was getting a solid 30psi right at the injector pump with the old lift pump with just a little needle wiggle that was in sync with the engine/lift pump cycle. I did design/install my own 'snubber' with a 0.002 in orfice (drilled through solder at the end of a fitting but the solder may have worn, enlarging the hole and reducing the 'snubber' effect). I would think that if the mods or extra filter was creating too much drag, I would not have seen the initial 30psi or now see the 40+psi peaks on the extreme needle bounce after replacing filters and lift pump, but a lower max pressure, and I haven't put that much milage or fuel through the truck since the mod's, and what WMO I did use is filtered down to 1 micron, so there hasn't been much trash put through the filters to load them down. I'm thinking the new Lift pump is what brought the max pressure back up, but for the life of me, can't figure out what causes the huge gauge needle bounce with each lift pump cycle unless the overflow valve is bad and is bleeding off a huge amount of fuel that the lift pump is trying to provide to where the pressure immediately bleeds off after each initial lift pump pulse? (which evidently isn't the case as I've just now replaced the overflow valve)

If anyone has experienced this and found out what causes this and how to fix, any feedback would be appreciated. I don't want to wind up damaging the injector pump...

Thanks!

 

The LP on the 12V is extremely violent and difficult to meter with an electric gauge, and since most people aren't crazy about a tube full of pressurized fuel running into their cabin, that's what most of us use. The "snubber" usually isn't enough to dampen the needle bounce, especially with a new LP. You have to get the sensor away from the fuel line like it would be with a mechanical gauge. I had a 2' rubber fuel line made with 1/8" NPT fittings to smooth out the pulses before the sensor picks them up. You can use a grease gun whip from Home Depot or someplace too, it's cheap and pre-made, but a lot stiffer material so it doesn't dampen as well. Regardless, if that isn't enough you'll need to add a needle valve on the pump side of the tube to restrict it further.
So, the arrangement would be: a needle valve screwed into whatever port in the fuel system you're taking your reading from, a 1-2' long tube screwed into the needle valve, the snubber screwed into the tube, and the sensor screwed into the snubber. BTW, this layout is straight from the engineer at ISSPRO, so I'm not just rednecking out my *** here LoL. Good luck.

 

hey J Bear,

Thanks for the comeback.

Ok, so it is "normal" for a gauge needle to wildly bounce back n forth from about 0-5psi to 40-50psi in sync with the LP piston cycles?

If so, then my original custom snuffer with a hole of about 0.002" must be worn through. I was worried that the 'violent' pressure bouncing wasn't good and concerned about injection pump damage. I guess I'll have to re-design my subber.

If it's so 'violent', would putting a check valve inline help keep the pressure up and more stable? If so, maybe something like a pressure absorber used on hot water baseboards to ease the hydrolic shocks, i.e. a vertical copper bulb with air in it where some fuel would flow in and pressurize (along with a one way check valve on the main line from the LP, then when the pulse drops off, the air pressure in the tank would push fuel back into the line to help smooth the violent pulse? Has anyone done anything like this? Is it needed, and would it extend the life of the LP?

Has anyone tried putting a check valve inline from the lift pump to the injection pump? I would think an IP would be a happy camper if the pressure was more steady than here today, gone in a flash then back again...

I guess I need to build an isolator of some sort as well .

Thanks again for some feedback on this issue. Will look for any further clarifications. I'm amazed I find so little discussions about this. Maybe most people never bother to actually monitor/test, until it's too late...

 

The bouncing needle is only bad for your gauge- it will wear out prematurely. It doesn't make any difference to the IP. Now, on a highly modified engine under load at high rpm the stock LP doesn't make enough pressure to keep the IP healthy, but that has nothing to do with bouncing it has to do with just plain not enough pressure. I would not try putting anything between the LP and IP. Isolate the sensor with a capilary tube of some sort, and add a needle valve if that's not enough. No matter what it will still bounce around a bit, but not like this. The only real solution is an electric LP. Go with a FAST or AirDog and the needle won't bounce around anymore.