Moonsoon on a 4age 20v... Now running!

[deltakatsu]

I've got a Monsoon recently plugged into my 4age 20v Silvertop. I can get it to crank, but there's no detonation happening. It previously started on the stock ECU.

 

I'm not quite sure where to go with this, so I am enclosing the log files. Thanks for any insight you have!

Toyota MR2 4age 20v Silvertop.pclx

PC Datalog - 2023-01-22 5;59;32 pm.llgx 2023_0122 TriggerScopeLog.llgx

[Vaughan]

Did you click capture on the triggerscope while it was cranking?

[deltakatsu]

14 minutes ago, Vaughan said:

Did you click capture on the triggerscope while it was cranking?

My bad, must've slipped on the capture button. I updated the OP with a correct scope file and a corresponding new logfile. Enclosed below is a screenshot of the scope.

 

[Vaughan]

Not getting enough Trigger 1 voltage, inspect your sensor for corrosion and gap, is it cranking at a decent speed or struggling?

[deltakatsu]

13 minutes ago, Vaughan said:

Not getting enough Trigger 1 voltage, inspect your sensor for corrosion and gap, is it cranking at a decent speed or struggling?

It's cranking at a good speed. The trigger is coming from the distributor, and it LOOKS okay to me. I cleaned the wheel of surface corrosion recently, and the gaps tested within spec.

[Adamw]

Yeah that trigger 1 voltage only reaching 0.1V is not usable.   Trigger 2 looks like it should be just ok.   

Can you do a triggerscope while cranking with the spark plugs out to see if the signal looks more acceptable with it cranking faster.

[koracing]

If you loosen the screws holding the module inside the distributor you can also adjust the gaps tighter which should raise the voltage of the signal.  Be careful though as it's a bit tricky getting all three trigger pickup gaps as nice as possible, and it's possible to make it worse than before if you're unsure what to do.

[deltakatsu]

19 hours ago, Adamw said:

Yeah that trigger 1 voltage only reaching 0.1V is not usable.   Trigger 2 looks like it should be just ok.   

Can you do a triggerscope while cranking with the spark plugs out to see if the signal looks more acceptable with it cranking faster.

Enclosed are some screenshots of the trigger scope. With the spark plugs out made no significant difference on the output voltage. I also tried cranking at two different distributor timings (turned all the way forward, and all the way back).

 

14 hours ago, koracing said:

If you loosen the screws holding the module inside the distributor you can also adjust the gaps tighter which should raise the voltage of the signal.  Be careful though as it's a bit tricky getting all three trigger pickup gaps as nice as possible, and it's possible to make it worse than before if you're unsure what to do.

This is probably the next thing to test, I suppose!

dizzy02_SparkPlugsIn_TriggerScopeLog.llgx Plugs in dizzy 01 - PC Datalog - 2023-01-23 9;09;09 pm.llgx SparkPlugsOut_TriggerScopeLog.llgx dizzy01_SparkPlugsIn_TriggerScopeLog.llgx

[deltakatsu]

And here's after attempting to regap the distro.

Regap_Dizzy - TriggerScopeLog.llgx

PC Datalog - 2023-01-23 9;56;01 pm.llgx

[Adamw]

It is worse now that is was originally.  What did you set the gap to?  The fact the voltage didnt increase with an increase in cranking speed almost hints there may be a wiring issue.  How has the ecu been wired - has it got a new custom engine loom, or using a modified stock loom or using a stock loom with an adapter harness?

If you have a multimeter, can you unplug the ecu and measure resistance between the below pins in the ecu loom plug (distributor still plugged in at the other end).  And report back something like this:

7 to 8 = ***ohms

7 to 9 = ***ohms

8 to 9 = ***ohms

[deltakatsu]

35 minutes ago, Adamw said:

It is worse now that is was originally.  What did you set the gap to?  The fact the voltage didnt increase with an increase in cranking speed almost hints there may be a wiring issue.  How has the ecu been wired - has it got a new custom engine loom, or using a modified stock loom or using a stock loom with an adapter harness?

If you have a multimeter, can you unplug the ecu and measure resistance between the below pins in the ecu loom plug (distributor still plugged in at the other end).  And report back something like this:

7 to 8 = ***ohms

7 to 9 = ***ohms

8 to 9 = ***ohms

I moved the magnets closer, approximating 0.2mm with a feeler gauge. I couldn't find a good guide on doing this but it seemed straight forward. I'm also ordering another used distributor just in case.

As for wiring, it's a patch harness from a vendor going to the OEM loom. It's a swap, but it ran for four years with no issues, so I'm fairly confident in the layout of the swap harness (but won't discount that it may have developed issues).

7 > 8 = 189.4 ohms

7 > 9 = 152.1 ohms

8 > 9 = 249.2 ohms

[koracing]

I would basically make the triggers touch the pickups and then move them away just enough to clear - maybe .5-1mm and see if it's any better. 

What wiring version is your 20v harness?  The blacktop 20V pinout is nearly identical pinout to the ST205X plugin ecu without the need for an adapter harness... just FYI.

When you say it ran for four years without issues - do you mean with and aftermarket ecu or with an OEM ecu?  The OEM ecu will possibly have better ability to read and use triggers specific to an engine since that ecu is only for that one engine versus an aftermarket ecu that tries to cater to every engine/trigger setup under the sun.

If you pull the distributor and spin it with a drill at closer to 1000rpm (disable injectors and ignition before you try this): what does the trigger scope look like doing that?

The last scope you posted looks kind of like when there's no trigger ground to the distributor.  What is resistance from pin 7 to ground (pin 25 or 34)?

[deltakatsu]

5 hours ago, koracing said:

I would basically make the triggers touch the pickups and then move them away just enough to clear - maybe .5-1mm and see if it's any better. 

What wiring version is your 20v harness?  The blacktop 20V pinout is nearly identical pinout to the ST205X plugin ecu without the need for an adapter harness... just FYI.

When you say it ran for four years without issues - do you mean with and aftermarket ecu or with an OEM ecu?  The OEM ecu will possibly have better ability to read and use triggers specific to an engine since that ecu is only for that one engine versus an aftermarket ecu that tries to cater to every engine/trigger setup under the sun.

If you pull the distributor and spin it with a drill at closer to 1000rpm (disable injectors and ignition before you try this): what does the trigger scope look like doing that?

The last scope you posted looks kind of like when there's no trigger ground to the distributor.  What is resistance from pin 7 to ground (pin 25 or 34)?

It's a Silvertop, set up to plug into the MR2's body harness and the Silvertop's OEM ECU. I was wondering what the OEM tolerances were for voltage, because 0.1V is still REALLY low. Is there no way to boost that signal? Though I've heard of other people getting 20v running on Link, so I'd imagine the problem is on my end.

I'll look into the drill tomorrow after my doctor's appointment. But is that going to be different from spinning the distributor with the spark plugs out?

As for the resistance: Assuming I'm looking at the pins correctly, I get open on both 25 and 34 from 7.

[Adamw]

Your resistance measurements on 7-9 and 7-8 yesterday look about right.  But the 8-9 one is a bit odd.  Since both sensors are connected together to the same ground wire, the resistance measured across 8-9 should have been the same as if 7/9 & 7/8 in series.  ie 190+150= ~340ohm.  Im not quite sure what that means, potentially some of the plastic insulation has broken down or there is water inside or something?  Or could have just been a bad measurement on one of them...

Your adjusted air gap of 0.2mm is about right.  You could potentially go a little lower depending how much runout and play there is in the mechanisim, but you shouldnt need too, you should well and truly be seeing a detectable waveform at that.   I have put a pic below showing the expected resistance and airgap just FYI.

So from those above observations I'm a little bit suspicious there may be some issue with one of these sensors but it is pretty rare for a reluctor sensor to fail (they are effectively just a coil of wire wrapped around a core), so I dont say this with a lot of confidence.  It will certainly be well worth trying another if you have one on the way.  

Your resistance test above between pin 7 and power ground is expected (assuming only the trigger sensors are grounded by pin 7), im sure the sensor ground is isolated from distributor body in the common toyota distributors so we shouldnt see continuity between 7 & 25/34. 

 

2 hours ago, deltakatsu said:

I was wondering what the OEM tolerances were for voltage, because 0.1V is still REALLY low. 

Yeah 0.1V is way too low for any ecu, at that level your signal isnt discernible from noise.  In fact your voltage is possibly even less than the scope shows, when in reluctor mode the trigger scope has to cover a large voltage range with only 8 bit resolution so I think the smallest increment it can show is only about 0.2V so some of this "signal" you are seeing is potentially just "rounding up" to the nearest 0.2V.   I dont think you will find an expected OEM voltage tolerance anywhere, but the toyota ECU's dont use a differential input or anything special so they arent going to be able to reliably detect anything much lower than 200mV which is about the same as Link.  

 

2 hours ago, deltakatsu said:

Is there no way to boost that signal?

Not electronically as that would also "boost" the noise at the same time.  VR sensor amplitude is influenced by 3 main factors - air gap, surface speed and number of "poles" (teeth).  So you have already experimented with the 2 you have control over.  Strangely your single tooth trig 2 sensor was originally putting out more voltage than the sensor with 24 teeth so that is part of the reason im a little suspicious of the sensor/s.  Trig 1 voltage is normally higher than trig 2 in this type of system.  There are other factors such as magnetism or permeability which are hard to isolate or test - but as an example I have seen crank trigger wheels "stop working" over a portion of its circumference after someone has mig welded something right beside it and magnetised it.  

 

2 hours ago, deltakatsu said:

Though I've heard of other people getting 20v running on Link, so I'd imagine the problem is on my end.

Yeah hundreds or probably thousands out there running with the stock dizzy, we have supported that engine since they were nearly new way back in G1 days.  Usually you need the cranking arming threshold relatively low but very rarely do they ever not make enough voltage to work.  In cases when they havent it is usually the air gap that is the problem.  

 

[deltakatsu]

6 hours ago, Adamw said:

but as an example I have seen crank trigger wheels "stop working" over a portion of its circumference after someone has mig welded something right beside it and magnetised it.  

Actually, my distributor was MIG welded about two years ago to fix a crack... It held on for a while though, but maybe that weakened it or something.

[koracing]

Is continuity good from one end of pin7 at the ecu to the distributor G- pin on the distributor plug?  I think if I'm translating the wiring diagram correctly the wire color should be white.

[deltakatsu]

8 hours ago, koracing said:

Is continuity good from one end of pin7 at the ecu to the distributor G- pin on the distributor plug?  I think if I'm translating the wiring diagram correctly the wire color should be white.

Strong beep, 0.0 to 0.1 ohms of resistance.

[deltakatsu]

It was the distributor...

 

Now to get the Innovate working with the Link! Regarding that, I saw an old post here saying the LC-2's brown and yellow wires output differently. If I am setting them both to output wideband in the Innovate software, is that enough, or do I have to splice both my gauge AND the ECU signal to the yellow wire?

I also wanted to verify that the Innovate should be body-grounded and not sensor grounded, because I'm getting conflicting opinions on that.

Thanks!

PC Datalog - 2023-01-30 5;01;47 pm.llgx Jan30_TriggerScopeLog.llgx

[koracing]

I would try it separately on the outputs and see if the ecu signal matches the gauge.  I've had one of these before that was supposed to output the right signal from both outputs, but did not, and I ended up splitting the signal between the gauge and ecu and that worked ok in that instance.  In either case you may have to mess with the input calibration to get it to read the same.

I would recommend grounding the LC-2 to go to a ground that the ecu also grounds to at a minimum or add a wire from where you ground the LC-2 to the sensor ground at the ecu to prevent analog output ground offset.  I know it's not too helpful given your current setup, but I tend to steer away from any wideband that doesn't either provide CAN connectivity or have a ground as part of the analog output to eliminate ground offset issues.

[deltakatsu]

Just an update on this: The LC-2 was grounded to the body right next to the ECU's ground. Both outputs from it were reading 5.5V with a multimeter, which should be outside its threshold. Tested against two different widebands and got the same results. I ordered a replacement controller preemptively. It's probably too late to go CAN for me, but what other gaugeless analog controllers should I have considered instead of the LC-2 for this application?

 

Another question: Is there a reason I'd have two Lambda inputs? It looks like they're being averaged together to my untrained eye. Or am I seeing it wrong, and I can disregard Lambda 1?

 

[essb00]

Pls attach your latest tune file so we can see how you configured the calibration table for the LC-2.

14point7 Spartan 2 is one good gaugeless analog wideband controller. The Spartan 3 has CAN.

If you only have 1 lambda controller connected (assigned as Lambda 1, nothing assigned on Lambda 2), then you should only look at Lambda 1.

[Adamw]

5 hours ago, deltakatsu said:

Another question: Is there a reason I'd have two Lambda inputs? It looks like they're being averaged together to my untrained eye. Or am I seeing it wrong, and I can disregard Lambda 1?

You can have up to 8 lambdas.  The Runtime "Lambda Avg" is the average of all connected lambda devices that are working, any that are reading 0 or not set up or reporting errors will not be included in the average calculation.  So if you only have Lambda 1 set up and working then the Lambda 1 and Lambda Avg runtimes should read exactly the same, you can use either one.  

I agree with essb00, if you really want to stick with an analog controller the spartan 2 would be my pick.  AEM X series is usually OK also, probably not quite as reliable as the spartan but generally good.  There are others like zeitronix that work well but with no separate analog ground I dont like to recommend them.  Innovate was junk in 1998 when it was first released as LC1, the only updates made to the LC2 was the removal of features and cheaper components, firmware and electronics in all their wideband controllers since is just a copy/paste of the LC1.  Even if you do get it to read correctly it will only be for a short time since they kill sensors like they are going out of fashion.  

[koracing]

I also like the spartan controllers but have found that right near stoich they seem to try to average or something and end up jumping just over or just under 1.0 Lambda with some built in buffer where like the AEM x series seems to read a bit more real time.  It's not enough to do anything that would prevent proper tuning - just something weird I've noticed.

[castillaricardo]

I also like the Spartan 3 controller. If CAN is not a wanted feature 14point7 also offers the Spartan 3 Lite that only outputs voltage and the Spartan 2 is also nice. The LC2 I had prior was off by a lot as I've found out.

[deltakatsu]

Since my truck took a turn for the worse, I need to get the MR2 running now, so I'll have to suck it up with the LC-2 until it dies, and then I'll get the Spartan or switch to CAN if that's feasible.

My main concern is I don't want a round gauge in the cabin. It's not a dedicated track/other car, so "good enough" on the wideband will drag me along for now.

In other news, I got the LC-2 talking to the Link, and the car idles. I'll run the Quick Tune so I can make it to the tuner shop next week and go for a dyno spin!