1990 1.6 mx5 low power?

Does anyone with an early 1.6 have problems with a lack of power below 4k rpm? When I first drove the car it felt like the brakes where dragging but I have since cleaned the brakes and replaced the pads and they were fine. I have also checked the crank pulley and there is a very slight wobble on idle but as soon as revs are applied this disappears. The carv seems to rev freely when in neutral but hesitates a bit under load. Has anyone had any experience with this, or any recommendation? I should also say I have done a diagnostic check and not faults where showing, and I changed the plugs and leads today with no big difference. I would be really grateful for any help,

There are a number of things it could be. 4k revs is about the point that the ECU starts ignoring sensors and does its own thing. So the likelihood is that one of them is sending bad responses. Could be temp sender, airflow meter, lambda/O2 sensor…my bet is this. There are test procedures for each in the workshop manual.

 I was going to suggest crank pulley, but you say you have checked it. Could try checking the timing. Could the cam belt have jumped a tooth.

Cheers guys, i have had a quick check of the timing but i will check again tomorrow to make sure.i did check the crank pully but i havnt really come across a problem like this so i’m unsure how much wobble is acceptible, is there any onther way of checking apart from taking it apart? i will also get my manual out and check the sensors, although would this not show up on the diagnostic check?

 

 When my air flow metre went I lost all power, and even dropping gears did not help maintain speed/power…good luck

Only if they are giving reading outside of range or not at all. They won’t log a code if they are giving incorrect readings (say a temp reading of 5C when it’s really 20C).

Lambda sensor (my best guess is it’s this) tend to sit giving a 0.5V reading all day long when they’re shot, but it only logs a code if it reads 0V or over 1V.

 

Ah i see, havnt been able to have a look today due to weather and other commitments but fingers crossed i can have a look tomorrow.

 

i have had a quick look at the lambda sensor and i can see 1 wire but can not find a connection. how do i tests the sensor? i have had a quick look in the manual and it just says how to replace it (i have the haynes repair manual). I did learn motorcycle mechanics at college but that being a few years ago i cant remember everything we were tought :frowning: if i dont use what i learn i tend to forget pretty quick.anyway sorry for all the questions guys just want to get this problem sorted now :slight_smile:

Trace it back and disconnect at the other end of the wire. Test the voltage beween the sensor and ground when the engine is running and at temperature (it generates its own voltage). Run the engine at about 3k and you should see 0.55V ish. Increasing revs quickly from there should see it climb immediately towards about 0.9V, reducing revs from 3k should see it drop towards 0.1V.

If it doesn’t change, or not very quickly/much, it’s probably shot. If in doubt, disconnect the chrome airpipe across the front of the engine to create a lean condition, then the sensor should be reading less than 0.1V…then squirt some carb cleaner or WD40 into the hole so it goes into the throttle and the sensor should read over 0.9V.

If you need a new one, don’t fall for Mazda prices. This type of Lambda/O2 sensor is very very common, and you can use any one that has 1 wire and the correct thread. There are new ‘universal 1-wire lambda sensors’ on ebay for just over a tenner which will work exactly the same as the Mazda ones for over £150.

 

ok have just done the test, i measured the voltage at idle and at 3k rpm and both times it ran about 0.8v, when above 3k rpm it went to aout 0.9v and below 3k rpm about 0.1v so raising and lowering rpm seems ok but the voltage for holding the throttle steady seems to be high. Am i correct in thinking this is showing as running rich as i did think it was using quite a bit of fuel. just cecked the timing again as i asnt sure it i did it properly, and it seems it is set to 14 degrees btdc. i understand this can be done to try and increase power in a way, but can this also cause problems on some cars? thanks for your help its really appreciated :slight_smile:

Yes, that’s way too high. It’s possibly your engine running way too rich (the exhaust would stink of unburnt petrol if so) or the sensor is duff.

Don’t forget to clear the fault codes as disconnecting it will have tripped a lambda sensor fault now.

 

ah that would explain a lot. well i suppose if the sensors are cheap enough i will get one and replace it and see what happens. its probably also worth putting the timing back to standard and see if that helps. at least now i know that the car is running rich and i can hopefully work from there now and fingers crossed i can have it sorted very soon. may need some more advice if i get stuck but for now thanks for you help :slight_smile:

If it’s not that, the next most likely thing is the ECU temp sender may be stuck sending a cold reading. That can be tested but it involves taking it off and measuring the resistance with it in water of varying temperatures…

yes i remember learning this test, looks awkward to get to so i’ll atempt it tomorrow after work 

i put the new lambda sensor in today, still measuring 0.8 - 0.9v at 3k rpm so it is getting a lot of fuel, for the coolant temperature sensor, does it require removing the coil? and what values should i see when testing?

 

That’s a shame, never easy!

Yeah coil pack off, sender wire off (scraped knuckles) sender out and into a beaker of coolant. Stick it in the freezer and come back to it and connect it to ohmeter, and measure resistance as it warms up:

-20C = 14.6-17.8 kohm

20C = 2.2-2.7 kohm

Then boil the kettle or stick it on the hob or whatever:

80C = 0.29-0.35 kohm

 

 

 

yes i was hoping that would sort it but just have to look abit deeper :slight_smile: will try the temp sensor soon, and report back then, cheers

Other than that I’m running out of ideas…there’s a bench test for the AFM involving checking resistances at various gate positions and the air thermosensor…it’s in the workshop manual, do you have it? If not I can find the test for you…

well hopefully testing the engine temp sensor tomorrow if i can get it out, 19mm deep socket i got doesnt fit  but not had enought time to try much else yet. i have a haynes repair manual (with the blue mk1 on the front) and i cant find much about the afm in that one, i think once i’ve done the checks on the remaining sensors (afm etc.) if that hasnt fixed it i might have to take it to a local garage see if they can find anything.