AEM Performance Electronics Forum

Transcription

1 AEM Performance Electronics Forum UNIVERSAL EMS => EMS-4 Universal Standalone => Topic started by: splice on February 27, 2012, 09:34:57 PM Title: toyota 4age 20v Post by: splice on February 27, 2012, 09:34:57 PM hello i could do with some help on my build. i have a toyota 4age 20v turbo and an aem ems 4 am using stock distributor and 2 coil packs from the toyota 3 rzfe for wasted spark i got the engine started and revs up to 3k rpm but then gets an ignition cut. the distributor has 24 teeth crank and 1 cam am stuck there any help would be appreciated please take a look at my cal. Post by: NS on February 28, 2012, 11:42:50 AM You're running an insane amount of dwell but your dwell max is limited to a half a tooth. I think a more realistic dwell max would be 3-4 teeth. Where'd you get your dwell settings from? I can't imagine that those coil packs would need more than 2-3ms of dwell. Post by: splice on March 05, 2012, 01:27:30 PM hello i change the previous settings on the map and still i couldn't rev pass 3k rpm i also reconfigure my map to ran a single coil with the dizzy and still i couldn't rev pass 3k rpm. just out of curiosity do i have to ground the shield from the 2 signal cables close to the dizzy? Am posting the 2 maps the tuner made 1 with wasted spark and 1 with single coil. plz take a look at my map am hope someone can help. thanks Post by: NS on March 05, 2012, 01:29:46 PM Yes, the shield should be drained to ground. Ground only the ECU end of the shield. 1/9

2 Post by: splice on March 05, 2012, 03:44:00 PM NS am using the aem 96" harness with relay kit isn't that grounded on the ecu side already? Post by: NS on March 05, 2012, 04:25:50 PM Sorry, I misread your post. No, do not ground the shield at the distributor. Yes, the -96 harness already has the shield grounded at the ECU end. Please post some logs showing the problem. Are you getting timing or sync errors? Post by: rbs14ryl on March 06, 2012, 02:27:10 AM I am helping splice with this car. The problem appears to be with the S Sync tooth. During smooth operation this number remains at 24 as we rev up to the 3000 ~4000 rpm range this value becomes erratic and bounces around at different values. Post by: NS on March 06, 2012, 01:02:28 PM Probably a noise issue. Are you doing any kind of filtering on the cam/crank signals? Post by: rbs14ryl on March 06, 2012, 02:55:15 PM No external filtering exists on the cam/crank connections other than the shielding inherent in the aems harness. I've read about the 5k resistor wired into the signal wires. Can you confirm the method in which the resistor should be wired in? Are they to be wired in parallel to each signal wire and ground? Any other noise filter recommendations would be appreciated. Post by: NS on March 06, 2012, 03:06:03 PM Eliminating noise can be a bit of a process. First, start with pull down resistors. All you're doing with a pull down resistor is allowing sensor ground to have some influence over the signal which reduces the amplitude of the signal as well as noise. Start with a higher resistance (weaker pull down) and go to a lower resistance (stronger pull down) until the problem is alleviated. I'd start with 10kOhm and then go to 5k, 2.2k, and 1k. If you still have noise issues at lower resistances then you may consider using a 2/9

3 low pass filter. Technically this is a parallel capacitor with a series resistor. The sensor's own resistance is the series resistor so all that's needed is a capacitor that ties the signal circuit to sensor ground. Start with a low capacitance like.001uf and work your way up (.01uF,.1uF, etc). The easiest way to do all of this is right at the sensor connector. Post by: splice on March 09, 2012, 09:41:53 AM hello NS i notice aem has pnp ECU for the toyota sc300 and 3sgte are the circuit configuration that much different as to the EMS4? Reason i ask is i examine the stock dizzy for both of these cars and internally they're the same as the 4age 20v. thanks for a response. Post by: NS on March 09, 2012, 11:29:44 AM Yes, they're different. Plug- and- play EMS's have obviously been configured to work on the application they're specified for. The EMS- 4 is a universal product so it needs to be configurable for any number of applications therefore it has minimal filtering on the cam/crank inputs. The filtering that might be needed actually has little to do with the internals of the distributor. It's outside factors that cause noise. For instance, specifically on the NA 2JZ cars you mentioned, the factory does not use shielded wire for the cam/crank signals and to make matters even worse, all six spark plug wires run directly over the cam/crank signal wires. This makes for an extremely noisy signal by time it gets to the EMS so filtering is required to eliminate this noise. I absolutely would not assume that what works for one application will work for another just because they're similar. Treat each individual application differently because they are different. You should start by looking for external noise inducers. For example, are the cam/crank signal wires being routed any where near the spark plug wires? If so, re- route them. How are your wiring conventions? Have you isolated noisy items (electric fan, fuel pump, etc) away from the cam/crank signal wires? Also, are you using shielded wire (I know you are but it always needs to be asked). I'd start with that pull down resistors that I mentioned and see where that gets you. Post by: TitaniumThirteen on March 09, 2012, 01:24:42 PM Have you tried adjusting Sync Ignore Above RPM? Say if you can't rev past 3000 RPM exactly, Try something like I ran EMS-4s on both 3S-GE Beams and 1NZ-FE engine, even though they share the same crank-cam pattern, the 3S-GE need to have this value adjusted to allow it to rev past 5000 RPM. It was very much like an ignition cut but there were no timing errors. I don't even know why but it worked :-P 3/9

4 Also, how did you wire the harness to the distributor? Post by: rbs14ryl on March 09, 2012, 01:42:04 PM I thought about changing the ignore above rpm, but was uncertain if ignoring the sync tooth could potentially hurt the engine. Thus far no timing errors have been noted while the car is misfiring and it does sound similar to an ignition cut. I was hoping we could solve the problem first before ignoring it. Good to know that you have had success with this method. thanks! Post by: NS on March 09, 2012, 02:07:43 PM I would NOT use Sync Ignore for this type of issue - ever. It was an option created to work around problems that came from a specific application years ago and the firmware has since been enhanced to fix any and all problems associated with the particular app. In fact, you can probably expect future firmware version to no longer have the Sync Ignore option available. It really should not be used at all in almost any instance. Post by: TitaniumThirteen on March 27, 2012, 11:09:39 AM Funny I have just witnessed the same problem in person with a friend's car yesterday. It's an AE86 with 4A- GE 20V reverse mounted. I recalled answering to this thread, discuss the issue with him and we decided to give a try. We had 2 EMS-4s and a 1040 box on hand. After the install and first startup it won't even go past 2000RPM. We re-route the dizzy wires away from the main loom, changed another dizzy and also tried adding 2.2 k resistors on the wire. After a couple hours we were able to rev it up to ~3k at most. Both EMS give same result. I have also notice an erratic on the crank teeth count value during the cut. We then splice the 1040 into the same harness, load the cal, start it up and the problem is simply gone... So far no solution has been found : Post by: NS on March 27, 2012, 11:51:22 AM Was the resistor added in series (in-line) or in parallel pulled down to sensor ground? Post by: NS on March 27, 2012, 12:08:53 PM Also, the 1040 box has a very strong pull down on the cam and crank inputs. 4/9

5 Post by: TitaniumThirteen on March 27, 2012, 01:31:56 PM It was pulled to its own ground, which also connects to the sensor ground and not the main. This is done only to the crank sensor though. Maybe it's noise on the cam signal causing extra sync? Because i notice an error only with the crank teeth count but there were no timing errors even if i keep on letting it cut. When it's normal it will count 23 teeth per cycle, when the cut occur the value is entirely random. Can i ask at what voltage level of noise will be able to trigger the signal conditioner circuit? Post by: NS on March 27, 2012, 03:51:16 PM Honestly it depends on the application. More specifically it depends on the total input circuit impedance and more specifically yet, it more or less depends on the sensor resistance. The voltage level where the comparator trips depends on current (limited internally) and resistance (internal resistance + external resistance). In some instances the trigger level can be fairly low (~200mV) so a good, clean, filtered signal may be necessary. It's nothing that can't be solved with regular filtering practices and good wiring conventions. Post by: NS on March 27, 2012, 04:06:06 PM What parameter is going random when the cut occurs? If it's S Sync Tooth or Sync Tooth then it could be indicative of noise on the cam input. And yes, I'd probably considering filtering the cam signal as well. The cam and crank signals are coming from the same place and being affected by the same noise sources aren't they? Post by: TitaniumThirteen on March 28, 2012, 12:11:51 AM Yup, it's sync tooth parameter that went random. So it's the cam signal then :-P Post by: NS on March 28, 2012, 12:07:56 PM FWIW, the 1040 has 1k pull downs on both the cam and the crank. Post by: splice on April 04, 2012, 08:15:11 AM hi thanks guys for all the help but in the end i got an aem 12 tooth crank wheel and hall effect sensors i still have one problem sometimes the ecu gives hell to start up it like you turn 5/9

6 the ignition switch on and the ecu wont boot up, there will be no priming of the pump and car wont start also my lap top wont access it need some help with that. thanks. Post by: TitaniumThirteen on April 04, 2012, 09:15:43 AM When it boots and connectable, give it a password ASAP :-D Post by: NS on April 04, 2012, 12:05:41 PM splice, what does your Rev/Mod sticker say? Post by: vprtech on April 07, 2012, 02:55:42 PM Maybe I missed it, but the CAS used on the 20V 4AGE is falling edge, and there is no way to switch the polarity. The EMS4 only supports rising edge MAG sensor signals. Curious, how did you fit a hall sensor with the 24-1 trigger to the engine, can you post a picture? Post by: Modular Wiring Solutions on April 11, 2012, 02:09:57 PM Having the same problem, 3sgte wont rev past 5k, with or without resistors. Wires are shielded individually and the harness is custom and has been reopened and checked for accuracy. All grounds are on only one point. Tried 2 distributors. Post by: NS on April 11, 2012, 02:39:52 PM What resistors did you use and what happens when rpm reaches 5000? What cam and crank pattern? Did you scope them to verify proper waveform or presence of noise? A log would be helpful. Post by: Modular Wiring Solutions on April 11, 2012, 02:51:21 PM I used from 5k down to 1k. Installed these from the crank signal to the return ground wire. In the distributor the return ground is common for both sensors so I tried each sensor with a resistor individually, both at the same time, etc. I scoped the signal when the distributor was outside and revd it to find out sensor polarity and everything checked out. I have not scoped it with the car running. 6/9

7 Will do that right now. Here's a datalog with a 1k resistor to sensor ground. Post by: NS on April 11, 2012, 03:16:49 PM Check out your Crank Tooth Period. Big spikes upward usually mean the EMS isn't seeing teeth (long period of time between teeth) while spikes downward usually mean the EMS is seeing extra teeth (shorter or implausible period of time between teeth). Sounds like you've got a flakey sensor/distributor or a bad connection. Like someone mentioned before, there's some Toyota distributor (from the 4AGE maybe?) that has a falling edge output that can't be inverted because there's a single signal ground. Are you sure that your distributor is outputting a rising edge? Also, I'd highly recommend using internal logging for logging stuff like this because of the higher capture rate. You should also record the Cam PER channel. Post by: Modular Wiring Solutions on April 11, 2012, 03:27:35 PM I'm trying to post the oscoped signal but it doesn't let me. The 1k resistor across cam and sensor ground seems to help up to 5800 but I don't see much difference on the signal itsel in the oscope. Post by: NS on April 11, 2012, 05:14:55 PM Might be easier to . Post by: Modular Wiring Solutions on April 11, 2012, 05:28:37 PM Send me your . I'm going to try a signal stabilizer from msd or haltech that converts it to square wave later on, but wanted to post solutions for people who might have this issue in the future. Post by: NS on April 11, 2012, 05:32:14 PM Click the little envelope icon under my screen name. Post by: splice on April 11, 2012, 10:41:11 PM sorry guys i have not been able to post a pic of how i empty out the dizzy and fitted a hall sensor 7/9

8 in it preparing for a drag race event this weekend as soon as i can i will post it Post by: Modular Wiring Solutions on April 15, 2012, 02:27:51 PM Update: I modified an AEM EPM to use the toyota 3S-GTE distributor base and now it revs clean all sync issues are gone. Videos and pics coming soon. Post by: PQatPIT on April 15, 2012, 04:57:54 PM One thing that might help in situations like this is reversing sensor wires and selecting the other edge than previously. I have noticed that it affects signal reading capability somewhat. And you have nothing to lose :) Post by: Modular Wiring Solutions on April 15, 2012, 11:51:21 PM Quote from : PQatPIT on April 15, 2012, 04:57:54 PM One thing that m ight help in situations like this is reversing sensor wires and selecting the other edge than previously. I have noticed that it affects signal reading capability som ewhat. And you have nothing to lose :) 1 - The ems-4 can only trigger on the rising edge of the vr sensor, selecting any other edge will cause random timing at high rpms. 2 - The toyota distributor has 2 signal wires with one ground, you just can't invert the wires. Post by: NS on April 16, 2012, 08:13:04 PM Quote from : Modular Wiring Solutions on April 15, 2012, 02:27:51 PM Update: I m odified an AEM EPM to use the toyota 3S-GTE distributor base and now it revs clean all sync issues are gone. Videos and pics com ing soon. Awesome. Glad to see that you got it all sorted now. Post by: splice on April 17, 2012, 12:24:37 PM modular am seriously considering doing this to my car hope you don't mind as i said i was very busy preparing for a drag event we had on the weekend didnt get a chance to post a pic of mine our best time was 12.1 we have a 4age 20v turbo in an ep82 starlet car breakdown on the 3rd run drop 2 valves cause we push the rev to 8650 rpm to help spool the turbo. i believe the turbo is too big, its a borg warner s257 the back has a 1.15 A/R am posting a log we did so hope you guys can give me some feedback 8/9

9 thanks Post by: Modular Wiring Solutions on April 17, 2012, 01:54:42 PM Just made 530hp to the wheel on a precision 6262 billet non-ball-bearing. There's no need to rev that motor more than 7500 rpm Powered by SMF SMF 2013, Sim ple Machines 9/9