Autopsy results on lift pump
#1
Registered User
Thread Starter
Join Date: Sep 2002
Location: In the middle of Weather Dry Creek Farm in Avilla, Arkansas
Posts: 161
Likes: 0
Received 1 Like
on
1 Post
Autopsy results on lift pump
Had my LP go Tango Uniform over the weekend. Saturday afternoon I noticed that the engine would studder around 2500 rpm. Pulled old filter and the pump wouldn't prime. Replaced the pump. Everything functions like it's supposed to. No codes were set.
Took the old pump to work and disected it.
The rotor had less than .0005" wear on the ends where it contacts the thrust plate/washer.
The vanes had about the same wear on the edges. All the contact surfaces looked good except for noticeable wear on the flat sides where they contacted the inside of the rotor. All this was done with a Mk. 1 Mod 2 eyeball.
I then took a camera I use for some of my optical alignments and looked even closer.
The vane flats had lots of scoring from contact with the inside of the rotor. Additionally, the rounded edge that acts as a seal with the inside of the housing were very unevenly worn and had signs of erosion that is similar the that caused by cavitation in large high speed pumps. While the wear patterns on the four vanes were similar, they were not the same either.
Looking at the rotor, I found the slots where the vanes ride were worn in a butterfly/bow tie pattern. This caused a substantial amount of "slack" in the clearance between the vanes and the rotor as the vanes moved outward along the eccentric shape of the pump housing. The pump housing itself is round, but the rotor is offset inside the housing causing the vanes to move in and out of the rotor following the outside wall. At certain points of travel, the vanes had as much as .0075" "slop" in the rotor.
The pump housing has a steel sleeve pressed in with 2 holes for the inlet and outlet. Wear patterns indicated that, at times, the vanes were only making contact with ~35% of the bearing area. Full contact with the bearing area was being made on less than 50% of the rotation.
I put the pump back together, did a bit of plumbing, hooked it up to a power supply, and a container of fuel and found that it would not prime itself. Once primed, it would flow ~32 oz. per minute but would not put out over 3 psi pressure when I blocked the output line.
If the vanes and rotor were hardened and the sleeve inside the pump housing was machined more precisely, I think the pump would hold up. However, the quality of the internal workmanship on this pump being what it is, I can see why the fail as often as they do. It wouldn't take much to make it a good pump.
Just my two cents worth.
Shortround out
Took the old pump to work and disected it.
The rotor had less than .0005" wear on the ends where it contacts the thrust plate/washer.
The vanes had about the same wear on the edges. All the contact surfaces looked good except for noticeable wear on the flat sides where they contacted the inside of the rotor. All this was done with a Mk. 1 Mod 2 eyeball.
I then took a camera I use for some of my optical alignments and looked even closer.
The vane flats had lots of scoring from contact with the inside of the rotor. Additionally, the rounded edge that acts as a seal with the inside of the housing were very unevenly worn and had signs of erosion that is similar the that caused by cavitation in large high speed pumps. While the wear patterns on the four vanes were similar, they were not the same either.
Looking at the rotor, I found the slots where the vanes ride were worn in a butterfly/bow tie pattern. This caused a substantial amount of "slack" in the clearance between the vanes and the rotor as the vanes moved outward along the eccentric shape of the pump housing. The pump housing itself is round, but the rotor is offset inside the housing causing the vanes to move in and out of the rotor following the outside wall. At certain points of travel, the vanes had as much as .0075" "slop" in the rotor.
The pump housing has a steel sleeve pressed in with 2 holes for the inlet and outlet. Wear patterns indicated that, at times, the vanes were only making contact with ~35% of the bearing area. Full contact with the bearing area was being made on less than 50% of the rotation.
I put the pump back together, did a bit of plumbing, hooked it up to a power supply, and a container of fuel and found that it would not prime itself. Once primed, it would flow ~32 oz. per minute but would not put out over 3 psi pressure when I blocked the output line.
If the vanes and rotor were hardened and the sleeve inside the pump housing was machined more precisely, I think the pump would hold up. However, the quality of the internal workmanship on this pump being what it is, I can see why the fail as often as they do. It wouldn't take much to make it a good pump.
Just my two cents worth.
Shortround out
#3
Chapter President
Join Date: Aug 2003
Location: Eagle. ID
Posts: 506
Likes: 0
Received 0 Likes
on
0 Posts
- nice post, good data, thanks for going the extra mile!
Now: How do we fix it? There are CERTAINLY enough of us Mechanical Engineers around with a hankerin' to tinker, that we can generate the intellectual horsepower required to design something better than this Well Documented POS!
Now: How do we fix it? There are CERTAINLY enough of us Mechanical Engineers around with a hankerin' to tinker, that we can generate the intellectual horsepower required to design something better than this Well Documented POS!
#4
Hardness?
Were you able to check the hardness of the materials? Generally when two components serve as a wear couple, one is somewhat harder than the other. You probably already knew that, I was just curious.
#6
Registered User
Thread Starter
Join Date: Sep 2002
Location: In the middle of Weather Dry Creek Farm in Avilla, Arkansas
Posts: 161
Likes: 0
Received 1 Like
on
1 Post
Not having access to a Rockwell tester, the hardness is hard to determine. However, using the file method, I found the rotor to be "soft" in that a single 2" stroke with a mill smooth file will definately remove material. The vanes are harder than woodpecker lips. The sleeve is somewhere in between.
It's interesting to note that there is a marked wear pattern at the point of maximum vane extension on the sleeve. An oscillation/chattering of the vanes is readily apparent. 180 degrees from that point, maximum vane contact is made with the sleeve. Even that is only for less than .250" of travel. Minimal vane contact is made before and after the inlet and outlet ports.
The housing appears to be a casting. IF the mold/pattern was not dimentionally correct and no machine work was performed in the sleeve area, and it appears that way, I can see where the sleeve would get distorted when pressed into the housing. This would cause the uneven wear on the sleeve and vanes.
Why not build the pump along the same lines as an oil pump? It can't be any more expensive. Even a direct drive pump would work. The electric pump doesn't have much of a current draw. Just 2 amps dry and a hair over 3 amps pumping. My guess is that a pump operating in it's optimal range wouldn't draw over 5 amps.
wornoutfromtheearlymorningbraincrunchingShortround out
It's interesting to note that there is a marked wear pattern at the point of maximum vane extension on the sleeve. An oscillation/chattering of the vanes is readily apparent. 180 degrees from that point, maximum vane contact is made with the sleeve. Even that is only for less than .250" of travel. Minimal vane contact is made before and after the inlet and outlet ports.
The housing appears to be a casting. IF the mold/pattern was not dimentionally correct and no machine work was performed in the sleeve area, and it appears that way, I can see where the sleeve would get distorted when pressed into the housing. This would cause the uneven wear on the sleeve and vanes.
Why not build the pump along the same lines as an oil pump? It can't be any more expensive. Even a direct drive pump would work. The electric pump doesn't have much of a current draw. Just 2 amps dry and a hair over 3 amps pumping. My guess is that a pump operating in it's optimal range wouldn't draw over 5 amps.
wornoutfromtheearlymorningbraincrunchingShortround out
#7
Registered User
Join Date: Sep 2005
Location: acworth, ga / camp lejeune, nc
Posts: 183
Likes: 0
Received 0 Likes
on
0 Posts
i wonder if it would be possible to adapt an oil pump over to take the place of the lift pump? this would be a very economical and effective way to pretty much eliminate teh lp failure. of course then theres the fass...
jeremy
jeremy
Trending Topics
#9
Adminstrator-ess
Have any of you 2nd gen guys taken a peek at Superduty's thread on lift pumps?
https://www.dieseltruckresource.com/...ad.php?t=78702
I think that would be an excellent solution for the VP44 equipped trucks too.
https://www.dieseltruckresource.com/...ad.php?t=78702
I think that would be an excellent solution for the VP44 equipped trucks too.
#11
Registered User
Join Date: Oct 2003
Location: Mason Neck Va
Posts: 149
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by farmer dave
Great information! Real interesting to know what exactly is causing the failures.
#12
Registered User
Thread Starter
Join Date: Sep 2002
Location: In the middle of Weather Dry Creek Farm in Avilla, Arkansas
Posts: 161
Likes: 0
Received 1 Like
on
1 Post
Not sure as to the brush issue. It appears that a certain amount of the fuel circulates around the LP motor to keep it cool. Don't know for sure since I haven't cut the housing in two, yet. The motor could be sealed or it could be a brushless design not requiring sealing from the fuel.
Either way, the motor runs fine. I still think that all the issues are on the pump side.
Shortround out
Either way, the motor runs fine. I still think that all the issues are on the pump side.
Shortround out
#13
Registered User
Thread Starter
Join Date: Sep 2002
Location: In the middle of Weather Dry Creek Farm in Avilla, Arkansas
Posts: 161
Likes: 0
Received 1 Like
on
1 Post
After the last post, I got to thinking about the motor area. I just finished futher disection and here is what I found.
Motor is cooled by fuel circulating through the housing.
Brushes and commutator were in good shape with ~35% wear.
Very interesting fact noted tho, there is a spring loaded ball check valve directly below the inlet side. There is another hole below the outlet side. Whether the ball is used as a check valve or pressure regulator isn't known. However, I would guess that it functions as both. I noted that the spring was large enough to allow the ball to actually slide inside the coils of the spring. It just rattled inside the housing. With the configuration of the spring/ball/seat combination, I'm leaning toward it being a pressure relief valve. (fuel comes into the inlet and from there into a chamber. Within the chamber is another hole going into the rotor/vane housing. The pump vanes then pick up the fuel and move it to the outlet hole. It goes from there to the IP but also into the motor housing. If the pressure gets too high, the ball will unseat and relieve the pressue back into the inlet (suction) side. If the spring cannot apply pressure to the ball, pressure cannot build.)
I'm thinking that the spring is too large for the ball. I wish I could have disected without tearing it up.
Ain't this weird.
scratchingabaldspotonthesideofmyheadShortround out
Motor is cooled by fuel circulating through the housing.
Brushes and commutator were in good shape with ~35% wear.
Very interesting fact noted tho, there is a spring loaded ball check valve directly below the inlet side. There is another hole below the outlet side. Whether the ball is used as a check valve or pressure regulator isn't known. However, I would guess that it functions as both. I noted that the spring was large enough to allow the ball to actually slide inside the coils of the spring. It just rattled inside the housing. With the configuration of the spring/ball/seat combination, I'm leaning toward it being a pressure relief valve. (fuel comes into the inlet and from there into a chamber. Within the chamber is another hole going into the rotor/vane housing. The pump vanes then pick up the fuel and move it to the outlet hole. It goes from there to the IP but also into the motor housing. If the pressure gets too high, the ball will unseat and relieve the pressue back into the inlet (suction) side. If the spring cannot apply pressure to the ball, pressure cannot build.)
I'm thinking that the spring is too large for the ball. I wish I could have disected without tearing it up.
Ain't this weird.
scratchingabaldspotonthesideofmyheadShortround out