Image Unavailable, Please Login The paint is now stripped away using a gel-like white gunk. It was easier than ever. Since the pump and accu are now in silver (old was black and dark green respectively), I think the brackets should be black as well. If someone wants to change this to original, that is possible. I will keep the old parts in a box.
Another issue that came up today was the torque needed for the union to the filter inlet. The manual gives a proposal for M12x1.5 (27Nm). I thought it was quite high for a steel to alu connection. To confirm what torque could be right I marked the angle of the union at the outlet, unscrewed it and cleaned off the Loctite. Then torqued it down to 15, 20, 23, 25, 27. The angle was pretty close at 25Nm and at 27Nm slightly larger angle but closer than at 25 ... so my conclusion is that the factory torque should be in that range.
Are you talking about the ringclamp on the lower half of the air box (that the air filter sits inside)? I've honestly never even used a torque wrench on that clamp; I just get it snug such that the airbox can't move/rotate.
No, it is the fuel filter inlet union. I think the union is steel but the filter is some aluminum alloy. Many alloys would have no problem with the applied torque but pure aluminum would.
Image Unavailable, Please Login A photo of the base coat for rust protection. Surfaces were cleaned, sanded with 600 grit paper and sligthly rounded edges.
Image Unavailable, Please Login This is the cap nut for the fuel pump. It was rusty and after sanding off the surface rust there is a ring visible in the middle. It is now hard to see. It has the same diameter as the swivel fuel pipe nuts inner diameter (fuel pipe opening). It seems the nut was manufactured from a swivel nut, but I am not sure.
You've sanded off the zinc passivation, so now it will probably rust again. Not to worry, you can get new ones from here: https://www.belmetric.com/banjo-bolt-caps-c-1041_929/
Yes, the plating is gone so I thought I could sand off the rust to see what was under. Apparently the cap nut is readily available here in Sweden too. More bits and pieces to buy ... it never gets boring. Anyway, I am reading some interesting articles on embrittlement of steel.
Image Unavailable, Please Login Image Unavailable, Please Login I received the new end nut for the fuel pump. This one is just a plain M12x1.5 nut for Bosch pumps. Measuring the thickness of the end cap it turns out to be just 0.5mm for the Ferrari nut and 3.5mm for the new nut. This one also has a larger ring spanner size. I did remove some rust on the old nut, maybe 0.5mm, so I guess it was about 1mm as original. I think this nut is worth checking out to avoid any surprises.
Just an FYI. The OEM for almost ALL our Peacock Gold gas & oil bits, lines, banjos, screws, bolts, etc., NO LONGER makes parts with that plating. FINISHED. THE END. Whatever you got in your paws is all there is. So it is now a 'Silver' Zinc finish and be happy or replate your own parts to get the Peacock Gold effect. Just peachy.
I was pleasantly surprised the new cap nut was much thicker material and better designed than the original. The "golden" plating is there too but I don't mind grey or silver. What I disliked about the original nut was the circular pattern in the cap which enabled moisture to find a way into the nut and cause corrosion. Bad design in my opinion. Merry Xmas!
Image Unavailable, Please Login Image Unavailable, Please Login I thought a few words about assembly of the fuel pump and fittings might be useful. I had some crow foot wrenches with good grip on more edges than the one pictured but these could not be used on the accumulator fitting. There was no space between the fittings. I had to buy standard two side wrenches for this job. It worked quite well, torqued this to 27Nm. I used both new copper washers and also some old ones which I annealed to soften them. There was one washer on the fuel filter I had to grind to a smaller outer dimension as well. The old metal pipe was reused. To not stress the pipe too much I tightened the fittings before final clamping of the pump and accu. It was then very easy to see if the clamping affected the pipe in any way.
Image Unavailable, Please Login The filler neck seems impossible with the upper clamp pointing inwards. However, it was easy to rotate the whole assembly to get some room to put the mini wrench in place. Image Unavailable, Please Login Next issue was the hose .. but I figured the neck rubber ring might help so I removed it first and then wiggled the neck off the hose through the hole, careful not to break the small vent lines.
Image Unavailable, Please Login Here the braided return line is installed. I then reused the metal pipe since it looks ok. The return hose was replaced with the new FKM rubber hose from Cohline. I decided to postpone the wur and fd rebuilds until later this year since time is limited. I guess it might be good to verify the fuel feed and injectors as a first step.
Image Unavailable, Please Login Now it is time to check the crossover hoses. It was easy to access the left one since there was no fuel pump assembly in the way. The old hoses were still soft but started to crack in some places. I removed the clamps and wiggled the hose inwards. There was some fuel left which made it easier to slide off the hose. The other side was easier since the pipe was then loose completely. Just move the pipe and hose to the left inwards. I have not yet figured out how to remove the pipe from the frame, though. I did not remove the drain plug. Instead I siphoned out the remaining fuel.
Image Unavailable, Please Login The crossover pipe was easy to remove if the glassfibre cover for the accelerator wire was removed. Installing the hoses had to be done with the pipe installed first. It was easy to slide the hose back and forth with a thin coat of vaseline on the pipe. I guess I could spend more time cleaning here, but it does not make much sense unless other parts were also cleaned. Maybe a later project. Now there are other garage maintenance I have to work on ..
Image Unavailable, Please Login Back to work on the 328 again ... I tested the new pump and pipes to the FD with good results. It looks great and control/system pressure are good. The injectors are not so good though, three leaking a drop in every few minutes and five with uneven spray patterns. Maybe cleaning might work but for how long? I think it is time for a full replacement. The cold start valve measured and tested just fine. The picture shows what contact I used to measure the valve circuit. With ignition off and the start valve wiring disconnected, the resistance between the contact (red probe) to ground is 60 ohms. This means the heating element in the thermo switch is ok. Connecting the start valve and with cool engine should read about 5 ohms which means the start valve wiring is ok and the thermo switch is closed. The start valve should spray fuel on next start attempt. Ignition on, but not in starter position. It is not necessary. Just bridge the ignition 12V (clamp 15) via a 7.5 A fuse (instrument lights is a good one) to the start valve circuit. The valve should spray fuel for a second or two depending on engine temperature. Then it stops and the circuit is opened by the thermo time switch.
Image Unavailable, Please Login I thought a delivery test might be good to perform. I used a digital scale with about 1% accuracy and a set of taco salsa glass jars ... not very professional looking, but working well. I think I have a photo of the setup somewhere. Each glass jar differ in weight, so each reading of the scale was a differential reading of the jar with fuel and without fuel. The WS does not mention acceptance criteria on this, but a Porsche manual suggest about 10% total deviation from the minimum fuel delivery. What I can conclude is the low delivery rate and medium delivery rate are within spec, but the full delivery rate is deviating too much. I have to think about the reasons for this, but it could be restrictions in the FD or injectors. Next step will be to swap some injectors around to see the effect of this. Also worth mentioning is the low delivery rate testing took too long and I became impatient and only got a resolution of 1:20. This is why the readings are 0%, 5% or 10%.
Image Unavailable, Please Login And here is the setup with glass jars. I have drilled a small hole in each lid to let air escape when injectors are operating. The original seal was used to let the injector sit firmly in the lid. Image Unavailable, Please Login The very simple injector tester is just a hose acting as a container, a cis pressure gauge (picture right below) and a swivel nut to attach the injector. The hose is filled with white spirit and then attached to an air compressor delivering more than 5 bars. The two gauges is for controlling max pressure (left upper gauge) and the gauge to the right is controlling actual pressure. Initial testing looks promising with a very precise injector opening control. I will post some results later.
The jars are not a bad idea but for fuel delivery you could just leave the rest of the system hooked up and in place, turn on the ignition and push the air flow plate down until you get a useful amount of fuel in the jars.
Yes, you are absolutely right. The two setups are separated. The fuel delivery is tested in car with the fuel pump and FD in place. The injector open/close and patterns are tested in the air compressor setup on the bench.
Injector testing was a bit messy and required a breathing mask and eye protection. This is highly recommended. The results were fairly consistent except for the close pressure which were depending on how quickly the pressure dropped. It seems odd but I guess that old injectors do not behave that well. Injector Open Pattern Close New 4.2 normal 4.0 F1 4.0 normal 3.2 F2 4.0 normal 3.0 F3 3.8 normal 2.8 F4 3.8 tilt 3.4 B1 4.0 normal 2.2 B2 3.8 tilt 3.4 B3 3.8 tilt 3.0 B4 3.9 tilt 3.4 My conclusion is that no injector passed on all three parameters. The new injector did well with a good pattern and 0.2 bar pressure drop from open to close.