Have you tried bypassing the fuel pump relay and resistor? It's supposed to bump up the voltage to the pump when you increase the load of the engine.

 

Not yet but i could measure the voltage at the pump tomorrow while driving around.

 

I believe it should be basically at 9 volts when cruising and then increase to 12 volts when you push the engine under load.

 

Check ground on the rear rotor housing again. Clean up the factory crimped on terminal and drop some solder on it. I had an issue like this with one of my old cars. I had made sure to check that ground, and was certain I got it nice and tight so I was looking over it when trying to diagnose the issue.
After trying many other things, I went back to that ground and sure enough it had loosened up. It was like hitting a brick wall at 5K.

 

I'm going to check that today. Is there anything else but the ECU grounded at that point? I'm asking because the ECU-grounds are all 4 ok. 0 Ohm towards the battery.

Or did you mean the main ground?

 

If I looked it up I could tell you. There are two wires in that crimped on ring connector though. It's not a bad Idea to put some dielectric grease on it after cleaning it, or some DE-OX or NoAlox to protect it from corrosion.

 

I just checked the wiring diagram and it says that somewhere the ground is split up and goes to the knock control unit. There might be a problem.

...Where the hell is the knock control unit?

 

Behind the passenger side kick panel on S4 LHD cars.

 

resistance and voltage levels can sometimes be misleading. once put under a load, bad connections then show.

 

I just got new stock injectors yesterday. Some other problems disappeared now but my strange bucking at 5000+ rpm still remains. I also changed the fuel pump back to an original one now. The trailing coil is new, too. I'm going to replace the leading one today. Got both of them very cheap so it was worth a try.

The fuel pump voltage seems to be ok, too. I'm running out of ideas soon.

 

Try bypassing the fuel pump relay located under the air box if you haven't done so already.

 

Ok I'm going to try that tomorrow. If that one is broken will I need to replace it or can I ignore it and leve it bypassed?

 

You can leave it bypassed as the net result would be the fuel pump would receive a constant 12 volts regardless of whether the car was under load or not. And another thing to think about is the fuel pump uses a certain amount of amps and a voltage reading would not tell you whether it was getting the proper amperage as a multi stranded wire with all but one strand intact would allow 12 volts to pass through but not the necessary amperage.

 

The resistor is bypassed now but nothing changed.

Now I've changed the spark plugs and wires, both coils, regrounded everything, got new injectors, changed the air filter and the fuel filter, changed the water temperature sensor and bypassed the fuel pump resistor. What the hell is causing it?

I think it all started when my upper intake was down for the first time to get the injectors out and cleaned. At that time I also regrounded the ECU wires because of that 3800rpm hesitation. Back then a mechanic also changed most of the vacuum lines.

By the way I heard about a "pill" which should be inside the vacuum line towards the pressure sensor. Should it really be there? Sounds like some valve to me but there's nothing in the line at the moment. Should I be worried about that? The mechanic might have thrown it away while he replaced the line if it ever was there.

 

You need the restrictor pill. You can make your own. Do an advanced search on the item and I'm sure you will find the necessary info to make one. W/o the pill it can cause hesitations but not sure at what specific rpm range.

 

I found some informations about the pill. It can cause fuel cuts under high pressure if it's not there. Looks like i will have to kill some people now...

Now i need to find the old vacuum line or a piece that fits into the new line and drill a 0,5mm hole through it. I would be the happiest guy on earth if that works.

 

Just one more question: on which end of the hose would the pill supposed to be?

 

Normally, it goes at the sensor side of the hose.

 

I made my own pill now. It seems like everything got much better now but there's still one problem with the car. Today I recognized that, if I stop the engine for about 20 min and try to start it again, I need 4-5 tries to get it running. It's the same with my old and with my new injectors so I guess it's not flooding.

Starting is like it spins, ignites 3 or 4 times on it's own but then stalls. It goes up to 1000 rpm but the immediately stalls. After the 4th or 5th try it starts like nothing ever was wrong.

After sitting those 20 mins I've seen today that the coolant temperature was a bit high according to the stock gauge but the temperature went down to normal in about 10 sec after I managed to start the engine.

Also another strange thing is that sometimes not all the check lights light up when I start the car. In fact only the ABS brake light lights up. Should I be worried about that?

 

I believe when a warm engine is started the FPR solenoid boots the fuel pressure at start up for a short period of time so maybe this is causing a flooding issue. Or maybe it's not working as supposed to. Or maybe the Water Thermosensor is not working properly and adding too much fuel at start up.

The idiot lights should light up w/key to on.

 

I can change the water thermosensor tomorrow. I allready got a new one from Mazda.

I changed the o2 sensor some weeks ago and found out the new one got a manufaturing error. I got another new one for free now but wasn't able to change it yet. Might that cause startup problems when the engine is hot?

 

I don't believe so, and what is a manufacturing error?

 

I meant the new sensor was damaged when I got it.

Bad news by the way: The hesitations came back this evening. It was a bit strange. They came back after the car was sitting for some hours . I warmed it up and they were there again. After warming it up I tried to rev it till 8000 rpm a few times and after doing that they were gone again and didn't come back till a few minutes after I started driving normally again (means no wot and no high rpms). It's like something gets clogged and I can blast it away by reving the engine up till 8000 at wot.