Good Afternoon All! My name is John and I am brand new to the Forum. I recently purchased a 2001 XKR. It keeps throwing the P1646 Code.

From what I have read so far, that appears to be associated with a Fuel Pump Relay 2 Malfunction. So far I have replaced the relays (on pumps 1 and 2) but have had no success. I also disconnected the Pump #1 relay and the car WILL NOT RUN on pump #2 alone.

I am assuming that Pump #2 needs to be replaced. Before I take out the fuel tank and replace the pump, is there anything else I should check? Also, while replacing the pumps, are there any other parts associated with the pump that should be addressed while the tank is out of the car?

Any insight is greatly appreciated!

Cheers!

 

John...Have you checked fuse #3 in the trunk fusebox? That is what I would look at next.

If it is not that, see if you have 12 volts between pins 1 and 2 of the pump #2 relay with a voltmeter with the ignition on.

I do not think you have a bad fuel pump.

 

With ignition on voltage across 1 and 2 is 13.5. With ignition off voltage between 2 and 3 is 12.5. Once the engine is started voltage between 2 and 3 goes to 0.

Fuses 3 and 7 are good.

Any other suggestions appreciated.

Thank you

 

Was 2 and 3 a typo...did you mean 1 and 2?

 

Unfortanately No. Once the ingnition is on, for some reason the voltage between 2 and 3 is lost. Almost as if the source power is for some reason taken away. I can't make sense of it.

Do you have any suggestions?

Thank you.

 

Actually what you are describing is what I believe to be normal operation. Pump number 2 does not turn on except when accelerating hard with the supercharger engaged.

Did you clear the code after replacing the relay?

 

On my 1999 XJR when the primary fuel pump stopped working it didn't trigger any codes. When I pulled the primary fuel pump relay it did run on the secondary pump. Pulling the triggered a code and the check engine light came on. The engine ran on the secondary pump but was limited to 3000 rpm.

 

I can clear the code but when I restart the car, it always comes back. I was under the impression that if you pull the pump #1 relay that #2 should kick in and run the car. It doesn't.

Could there be a problem with the ECM? Could there be a short? WhiteXKR does not think that it is a bad pump (which would be good). I can't make sense of why the check engine light keeps coming back and/or why the car wont run without the pump #1 relay. The car WILL run with the #2 relay out but under the restricted performance (3000 RPM)

Also, maybe it's my imaginiation, but SOMETIMES under hard acceleration, the car seems to be lacking a little bit of the pep it previously had.

Any thoughts on where I should venture next?

Thank you.

 

In the process of trying to help you, I think I stumbled on the fact that my secondary pump is not running!

Unlike you though, I have no codes.

Check continuty in your secondary pump: First verify fuse 22 in the trunk fusebox is good. Then with the secondary pump relay removed, check the resistance from pin 5 of the relay socket to ground (with ignition off).

Then consider that there has been conflicting documentation on P1646. Once would think it should be the fuel pump 2 relay on SC cars, but I am not so sure now. See https://www.jaguarforums.com/forum/x...nfusion-57005/

 

Try pulling relay number 2 and starting the car. See if you get P1671. That is the code I get, and it is also listed in some charts as a fuel pump relay 2 failure.

If this is the case, I retract my earlier statement, and think you have a fuel pump 2 failure like I have (which generates no codes), but is demonstrated by not being able to run on fuel pump 2 when fuel pump 1 is intentionally disabled.

I also think the P1646 is probably an upstream oxygen sensor (or connector/wire to the oxygen sensor) failure on Bank A.

I think the codes were switched sometime in mid-2001...very confusing.

 

I agree. The incidence of P1646 is reported much more often than a Relay Coil Circuit failure would typically occur.

Probably 99% of all Fuel Pump Failures are due to bad Fuel pumps or the Harnesses feeding them. Or (less likely, but possible) a bad CONTACT in the Relay. These failures DO NOT (directly) generate a trouble code.

 

I've been watching this thread and it's a very interesting puzzle. Several issues have been mentioned and the more I dig, the more unusual it seems.


I think there has been great advice here from experienced Jag owners. Very nice.

As a mechanic, I like to prove and disprove theories with tests before I sell a part. On my own cars, I do the same to keep from wasting money and to learn.

Do any of the previous posters (or anyone else) think it would be a good idea to use a fuel pressure guage to check the performance of each pump? Perhaps you could remove both f/p relays and jump the fuel pump feed terminals one relay socket at a time and monitor fuel pressure. Terminals 3 and 5, on my wiring schematic are shown as the correct terminals to jump across on both f/p relays. This may save you the cost of a fuel pump replacement. A fused jumper would be a good idea.

I did see that on a 2000 XKR P1646 indicates a HO2S bank A upstream fault code. On a 2001 and 2002 it indicates a F/P relay #2 control circuit problem. On a 2000, P1671 indicates a F/P relay #2 control circuit code. As was mentioned, the code changes may have happened mid model year and this may be why one 2001 XKR shows a P1671 and another shows P1646 with the same relay removed and started or while running.

To make things even more confusing, the scematic I'm looking at shows F/P relay #2 feeding pump #1 and F/P relay #1 feeding pump 2.

John, you mentioned that the voltage between terminals 2 and 3 went to 0v when the engine was started. As was mentioned this may be normal. Terminal 3 is shown as battery feed on both relays. Terminal 2 is shown as the ecu control circuit on both relays. This is where the ecu grounds the relay coil to run the pump

With the ignition off, terminal 2 has no power. It will read 12v between 2-3 because terminal 3 is battery postive and because 2 is de-energized, it will have some negative potential. With the ignition on or running, terminal 2 will have power unless the ecu grounds the circuit to energize the pump. The fact that you have 0v between 2-3 may be because both terminals have power.

Unfortunately, I have not been able to find a description of the fuel pump operating strategy. I do remember several other manufacturers strategy. Usually the dual pumps are used to provide a boost of fuel volume under heavy loads. Sometimes during cold start as well. Under low load conditions one pump will be used and when a heavy load is present the second pump will be activated. The second pump may be used as a "limp home" pump. Hopefully, someone can shed some light on this.

I think you need to establish if the pumps work with a pressure test.

Then, perhaps you can rig a voltmeter to give you readings from the control circuit of pump relay #2. Monitor the reading under heavy loads or perhaps by snapping the throttle.

From what I can tell, relay 1 is inside of the fuse relay box and has a 20 amp fuse in it. Relay 2 is next to the fuse bos and has no fuse in it. I can't confirm this because I'm looking at a diagram and not the car.

If the pumps and relays work ok, the upstream bank A H02S circuit is a good next step.

Any thought on this?


I have a headache now. Good luck.

 

Sorry if you have already mentioned it. But when you turn on the ignition, do you hear the fuel pump running for about 2 to 3 seconds?

 

You should test both pumps, one at a time, by jumping the relay sockets as other posters have described. From what you have said it sounds like the primary pump is working and the secondary is not. I'd still check them both anyway. If you can't hear both pumps running inside the tank, more troubleshooting is needed for whichever pump isn't running. You can connect +12v and ground test leads directly to the connector at the top of the tank. On my 1999 XJR it's a white 4-pin connector. The 2 pins on the right are the primary pump and the 2 on the left are the secondary. On both sides + is the lower pin and - is the upper. If you can hear both pumps running, that's great. You still might want to test the fuel pressure if you have a gauge. If either pump doesn't run by jumping power directly to the connector at the top of the tank, then you need to pull the tank. At that point it could be a bad pump or sometimes the connectors inside the tank can overheat and lose contact.

 

I want to thank everyone for thier valuable input. As a new member I am blown away at the community of Jaguar Enthusiasts on this site, truly brilliant. I have had a Jag for over 10 years, and like i tell everyone, nothing runs like a Jaguar. Anyhow, back on topic.

I have pulled the relay #2 and I do get the 1671 code for the missing relay, in addition to the restricted performance. I might be silly, but the performance of my car seems streaky. This morning I got it to go 0-60 in 6.0 seconds. Which I would think is about as good as it should be. Other times it doesn't feel the same. Almost as if the pump #2 sometimes isn't working and giving enough fuel.

From experience, could this problem at all be associated with a fault in the ECM itself?

Also, I do not own a fuel gauge. Should I buy one? If i do, where would the reccomended connection place be? Or can I test the pump by jumping current directly to it or across the relay terminals and listening? When the term 'jump' was used, do I take a fused wire from 1 to 3 and 2 to 5 with ingnition on? Is a 20A fuse okay?

The relays do not appear to be internally fused. They have both been replaced with replacement 40A Ford ones (From the Jag dealer $$$).

If you don't mind, please clairfy these queries for the simply minded. Is there anything else I should be careful of? Still new to this, but eager to learn.
Cheers!

 

Two more things, resistance between 5 and ground is open. Measurable resistance of 10.4 on the 200k setting between 2 and ground. 19.4 between 1 and ground, -115 between 3 and ground.

Fuse 22 is good.

Thank you

 

You can remove a Fuel Pump Relay and jumper the Socket Terminals to energize the Pump directly with 12 volts.

Using a short Jumper Wire, insert the wire into the two socket terminals ACROSS from each other that are also PERPENDICULAR to each other.

The terminals that are across from each other and are Parallel to each other are the feed wires to the Relay Coil. Do not Jumper these Terminals.

With the Engine not running and in a quiet area, you can hear the Pump Running in the Tank.

If no sound is heard, the Pump or the wiring harness to the pump is bad. The Harness inside the Tank can overheat and burn open. In either case, the Pump Assembly must be removed from the Tank for Replacement.

When I had Fuel Pump trouble, I jumped the relay socket and some sparking occurred but the Pump did not run. Also, the Fuse did NOT blow during the test. This occurred on one of my pumps. The other one was also bad but no current flowed. That Pump was Open Circuit. Both Pumps had to be Replaced.

 

+1 to Pauls advice. The relay#2 socket pin numbers to short to manually try to run the pump are 3 and 5. DO NOT experiment with shorting anything else or you might damage the ECU.

5 to ground being open is also an almost sure sign of a bad pump #2 (or a burned up link lead to the pump in the tank). This is exactly what I found on my car, and my pump #2 is definitely inoperative.

It is possible the pump is intermittant, and that may account for performance differences.

 

Not sure why you are focussing on the fuel pump, especially not now as you have proven that on your car the P1646 is related to the O2 sensor on Bank A. What is actually the build date?

 

He almost definitely has an O2 sensor problem that needs resolution, but ALSO has shown his car will not run on fuel pump #2 when #1 is disabled, and his fuel pump #2 shown infinite resistance.