This is a reference page I created on tracking down a turbo control problem I had. Its mostly for my own reference if I ever have the same problem again (or see it on a friend's car). But I thought others might find it useful.
The following are posts I made on the Rx7Forum:
|Weird Boost Issue - 8/24/2004|
|My apologies if this has been covered before. I've searched the forum and found about three posts describing my exact issue, but there were never any follow-ups explaining a fix.
Anyway, I've got a weird problem and wanted to see anyone has run into it before:
Details: PFC, Downpipe, Hi-Flow Cat, RB Dual Cat-Back, M2 Intake, new plugs, good wires, vac hose job, running 12 psi. I just installed the PFC and Steve Kan tuned it for me (great work!). I didn't have this problem with the M2 Stage III ECU I had before. I'm hoping its just that the PFC is more sensitive to something or there is a setting I've missed.
It does this in any gear, but for example, if I punch it in 2nd gear at 2500 rpm, I get full boost quickly, it transitions quickly to the secondary at around 4200 or so, and makes great power to redline... everything works great. However, if I let out of the throttle and coast back down to a lower rpm, say 4000, and punch it again, the boost builds very slowly and the exhaust note changes (sounds different, but hard to explain... different pitch). It will do this until the rpms drop below 2800 (either by coasting down in gear, or just putting the clutch in). I can coast all day (refreshing the vaccum supply), but I won't get quick boost back until the rpms drops below 2800...
Looking at the switch/relay readout on the commander, when I pass the transition point under heavy throttle, the CCN indicator goes out and the TCN comes on. It stays this way until the rpms drop below 2800, when it switches back and I can get quick boost again. Basically, I can get normal turbo operation once with WOT under a gear, and then I have to clutch in and let the rpms drop below 2800 before it will work right again. If I don't do that and then go WOT, it will take two or three thousand rpm to built back up to full boost.
I've tried replacing the turbo control solenoid (mounted on the ACV) and I also replaced the two solenoids mounted on the front of the upper intake (wastegate solenoid and pre-control) since they were easy to get to and those made no differences. I've also confirmed that all my checkvalves are good and my vacuum and pressure tanks are holding vac/press overnight.
Have you ever seen that behavior before. Any ideas?
From the exhaust sound change, I'm wondering if the wastegate or the other flapper door in the turbo manifold (prespool door) isn't functioning properly, but I can't figure out why it would work normally one time and then be screwed up until the rpms drop below 2800.
Is there any chance my PFC is bad? Everything else seems to work just great.
I'm going to take the upper intake off and check all the lines for kinks. I think I'll also replace the other turbo control solenoid (the one in the rack) while I'm in there.
Thanks for your time!
|Thought I'd make an update... I think I'm closing in on it.
I took the upper intake off and decided to do the block-off plates while I was in there. I kept the air control valve, since I still have a cat. But I eliminated the AWS, Double Throttle, and EGR. This let me take two solenoids out of the rack and simplify things a little.
I tested the turbo control actuator (big one on the bottom of the engine that uses pressure and vacuum). It worked fine. Then I checked the turbo control solenoid in the rack (3rd from the front). This one runs the pressure side of the TC actuator. It was sticking and acting weird. Sometimes it would stick open, sometimes closed. I thought I had it. Swapped in a good spare, put the car back together and went for a test drive.
It was a little worse than before
Now I get proper operation once and then pretty much no boost below 4500 after that. Letting the RPM drop below 3k doesn't reset it anymore. It'll make a pound or two under 4500 and then at the transition point, whap, full boost and power to redline. A couple of times while I was at wide open throttle at about 3k or so, it made 2 psi for a couple of seconds and then jumped right up to full boost, so I'm convinced that something is sticking (solenoid or actuator) and then popping free.
Also, there is a definite change in the exhaust note when the car is trying to make boost on the primary turbo (under 4500 rpm).
So, I started looking around and testing everything. I put vacuum on the A chamber of the charge control actuator (moves the valve in the Y-Pipe). It was slow to close when I removed the vacuum.... sometimes it would not close at all for about 30 seconds and sometime it would close halfway and after a little while it would close the rest of the way. Ah ha, I thought, thats it.
I removed the y-pipe and took the CCA apart. The actuator itself was fine, but the valve was gummed up a little and was causing it to be sticky. I cleaned and oiled everything and it worked smoothly. I put everything back together and went for another test drive. No change.
At this point, I've replaced both Turbo Control solenoids, the wastegate solenoid and the turbo pre-control solenoid. I checked again to make sure my vacuum and pressure tanks were holding pressure (they'll both hold overnight) and I've tested all the check valves multiple times with a mity-vac.
I've tested the operation of the CCA and done the key-on, key-off test and it passed. The only turbo-related solenoid I haven't replaced is the CCA solenoid. So I'm going to dig in and swap it out. I'm also going to check/re-test the solenoids that I've replaced so far, just to make sure I didn't get a bad replacement.
I'm convinced that something is sticking. If a new CCA solenoid doesn't fix it. I'm going to put my bullet-cam in the engine compartment and video the CCA, wastegate, pre-control, and turbo-control actuators and compare them to a healthy car. From the sound the exhaust makes I figure one of those is the problem. Also, my pre-control and wastegate actuators are pretty new (new turbos from Mazda a couple of years ago), so I haven't checked those yet, but they're next on my list.
|Ok, today I replaced the Charge Control Solenoid. While I was in there, I went ahead and checked (using the shop manual's procedure) every solenoid again. They're all working just fine. My car behaves the same if its cold, so I don't think any of the solenoids are heat soaking and causing problems.
I put voltage to the turbo control solenoid (vacuum side) and the TCA operated smoothly. Then I did the same to the pressure side turbo control solenoid and the TCA operated smoothly. Then I removed the lines from the turbo control solenoids and used the Mity-Vac to actuate the TCA many times. It never showed signs of sticking.
I did the same (mity-vac test) on the Turbo Pre-Control and Wastegate actuators. I actuated them about 10 times each and they always operated smoothly.
I pulled the pill lines on the turbos off to check them while I was in there. I forgot that my turbos (being new from Mazda a couple of years ago) have the restriction integrated into the turbo housings, so I don't have any pills at all. I thought one might have gotten rotated or blocked somehow.
I put it all back together and took it for a drive. The problem seems to be getting slightly worse. I still get one good pull and then NO boost below 4500 after that. Letting it drop below 3000 or putting the clutch in doesn't always help.... and I'm still getting a droning noise in the exhaust system when the car is making zero boost below 4500.
When its making no boost below 4500, If I go WOT and then step off the throttle, and repeat about three times, it will eventually "break free" and make full boost again, until I run it past the transition once and the whole process starts over.
I think I've pretty much eliminated any sort of mechanical problem, so I'm going to attack the electrical next. Oh, I still need to actually pull the actuators off the wategate, pre-control, and turbo control doors and check them for adjustment and check them for smooth movement.... but I don't expect to find anything since it will work right for one pull.
I didn't get to try the camera, that's next, but I'm going to have to make a custom mount for it under the car. If I can tell for sure which actuator is staying open, compared to a healthy car, I can concentrate on it and check the eletrical and re-check the mechanical.
|Ok, I worked on it again today and figured out what the problem is. But I still haven't fixed it.
Started out by re-checking the check valves for the vacuum and pressure tanks. In addition to vacuum testing them, as I'd done before, I also applied pressure to them. One (the vacuum tank valve) could only withstand about 10 psi of pressure, so I replaced it with a new one that would take 20 psi.... but I didn't think that was the problem and turns out it wasn't.
I checked the vent lines for the wastegate and pre-control solenoids (the lower nipples on each) to make sure they weren't blocked and holding pressure. They were fine.
I also checked the vent filters (the little round filters that are attached to the solenoids) on both turbo control solenoids to make sure they weren't blocked. They checked out ok.
Then I checked the adjustment on the Pre-Control and Turbo Control actuator rods. Both were OK.
Now here is where it gets interesting.
I bought a $20 web cam from CompUSA. It is crappy quality, but it has a built in white LED light and is USB and capable of 30 frames per second. The light was a plus (in a dark engine compartment) and I didn't want to risk my expensive ChaseCam if I didn't have to.... turns out that was a good idea (read on).
I placed the camera where I could see the Charge Control actuator and went for a test drive. It behaved exactly as it should, actuating at 4500 and turning off at 3000. I did quite a few runs and it worked perfectly every time.
Next, I placed the camera where I could see the Turbo Control actuator. I did a WOT run in second gear and the TC actuated at 4500 (as it should). Then I coasted back down to about 2800... TC stayed open, 2500 rpm, still open, put the clutch in and stopped. About the time I stopped it finally shut. I did several test runs and it behaved inconsistently every time. Sometimes it would slowly open after dropping below 3000, sometimes it wouldn't open at all, sometimes it wouldn't open until I hit 3500 or 4000 on the way back up during another WOT run.
Then the camera melted I figured there was enough airflow down there to keep it from melting (guess I should have used that aluminum foil tip, Dgeesaman)
So the TC is the problem.
I hadn't checked the TC actuator while the car was hot ('cause I don't like being burned ). So I checked it right after a few WOT runs. The vacuum and pressure lines for the TC run right behind the air pump and are pretty easy to get to there. I hooked my mity-vac up there and applied vacuum or pressure to the TC actuator and it worked perfectly every time. I thought about going ahead and replacing the TC actuator with a spare, but thought I would run another test first.
So, then I tee'd a vac/pressure gauge into the vacuum side of the TC actuator and went for a drive. It worked perfectly. At 4500 I would get vacuum on the gauge and at 3000, poof, the vacuum gauge dropped to zero.
Went back home and moved the gauge over to the pressure side of the TC actuator. The pressure side of the TC line is showing the same weird behavior that the video camera revealed. The pressure never left the line at 3000. Sometimes it stayed pressurized (TC open) through several runs, causing zero boost below 4500. Sometimes it releases after a few seconds, etc.
So.... something is keeping the pressure from leaving the TC actuator.
It can't be wiring because: Both turbo control solenoids are wired together to a single output on the ECU. Recall that the ECU grounds the solenoids to activate them. If one solenoid connection had a frayed wire that was grounding intermittently, it would also activate the other solenoid, which isn't happening. It can't be an ECU problem for the same reason. The ECU only has one wire to ground for both TC solenoids, so if it was keeping the solenoids activated, I would have also seen a problem in the vacuum TC solenoid.
I can't think of anything the TC actuator could do that would keep pressure in the line. When the pressure TC solenoid shuts off, it basically switches the TC actuator (pressure side) from being connected to the pressure tank to being connected to a vent to atmosphere (thus dumping the pressure from the line and shutting the TC door).
So, that leaves me with two possibilities. A) My replacement TC solenoid (pressure side) is also bad or B) I have a restriction in the line between the TC actuator and the TC solenoid.
I've tested the pressure TC solenoid several times (and replaced it once), but not when its really good and hot. It could be failing when it gets warmed up.
But I'm betting on option B. I think there is gunk (or something) in one of the steel pipes that connects the TC solenoid to the actuator. That would also explain why so many people have had this problem and replacing everything (solenoids, actuators, etc.) doesn't fix it.
I'll find out tomorrow and let you guys know. I'm going to check the steel pipes and clean them, or just bypass them completely with a long piece of hose. I will probably also replace the pressure side TC solenoid with a 3rd spare that I have.
Also, I can post those videos I made if anyone wants to see them. They are really poor quality, but all I needed to see was the actuator arm move.
- Dave Disney
|I fixed mine too.
It was the pressure side Turbo Control solenoid. The one I replaced it with was also bad.
Both the original solenoid and the replacment still pass Mazda's check procedure (in the shop manual) with flying colors.
But here is what's happening. There is a spring in the solenoid that returns the valve back to its initial position when the power source is removed from the solenoid. Over time, the heat weakens that spring and the pressure in the line between the solenoid and the turbo control valve holds the solenoid open (the spring cannot overcome the pressure). So, the pressure can't escape and the turbo control actuator is held open when it should not be.
The solenoid I had in the car couldn't take more than about 9 psi without sticking. The original solenoid in the car couldn't take more than about 11 (I'm running 12.5). I tested several spares I had and finally found one that was happy even at 15 psi. I put it in the car and went for a test drive..... car now performs perfectly and life is good
I can post a detailed test procedure for the solenoid valve if anyone is interested. You basically need to put 12v and a ground to the solenoid, then put pressure (about 12 psi to start) on the nipple that would go to the actuator and then remove the 12v and see if the solenoid lets the pressure in the line vent. If it doesn't (or is very slow doing it), then you have a problem. Let me know if anyone wants a nice how-to with pics. You have to test it as it is used in the car, the blow through test the shop manual outlines will not work with this solenoid. A very bad solenoid can still pass that test.
Thanks for all the troubleshooting help guys. My advice, if you're having the problems described above would be to buy a NEW pressure side Turbo Control solenoid.
- Dave Disney
Ok, now for a couple of notes. The turbo system goes into twin mode when you pass the transition point (about 4500 rpm at WOT, later for partial throttle). It does not return to single mode until you go below 3000 rpm. So, if you go past the transition and then coast down to 3100 rpm or so and floor it again, it will be a little slow building boost because both turbos are still on-line. It is supposed to work that way. The problem I was having was the turbo control valve was staying open when it shouldn't be, so sometimes I would have zero boost until 4500 (because most of the exhaust flow was going through the secondary turbo and out the charge relief valve) and other times primary boost was really slow to build up (maybe 4 or 5 psi by the time I hit 4500, then full boost).
As you can read above, it was a bad turbo control solenoid (pressure side, in the solenoid rack, 3rd back). You can also read about all the diagnostics I did. Basically, my first solenoid was bad and the spare solenoid I installed was also bad. Yet they both passed the test in Mazda's shop manual:
Which basically says to make sure air flows ONLY between B & C when no voltage is applied and that air flows ONLY between A & B when 12 volts is applied.
As I said before, both bad solenoids passed that test with flying colors. That test should work fine for the solenoids that only see vacuum. The problem here was that the springs in the solenoids were weak and pressure in the line between the solenoid and the turbo control actuator was holding the solenoid closed after the solenoid had been triggered once.
Here is a test that will let you check the solenoid for operation under boost:
click for larger image
You need something to simulate an actuator of some sort. I used a spare Charge Control Valve that I had lying around (even though this test was for the Turbo Control). Be sure your actuator uses pressure and not vacuum. If you don't have anything like that around, about 12 feet of vacuum line with a plug in the end should work (simulates the storage function of the actuator). You also need to tee in a boost gauge so you know how much pressure you're adding. If you pump too much pressure in, the solenoid will always stick open. Hook the lines up as shown above and make sure you're using the pressure output of the Mity-Vac and not the vaccum. Put 12v and ground to the solenoid (it should click) and pump about 12 psi into the system. The pressure should hold and not leak out for at least a minute or two. Remove the 12v and the pressure should immediately be released. If it is not released or bleeds out very slowly, you have a bad solenoid. A good solenoid should be able to handle at least 12 or 13 psi. I went through a pile of used ones until I found one that would withstand 15 psi (I only run 12.5). A new solenoid should be able to handle 15+.
Here is a video of a solenoid that fails the test (the evil, bad, naughty solenoid from my car): Bad Solenoid Video
Here is a video of a good solenoid: Good Solenoid Video
Here are the engine compartment videos I took while trying to diagnose the problem:
Charge Control Actuator Video
You'll note that the Charge Control Actuator performs as it should, opening at 4500 and then closing at 3000.
Turbo Control Actuator Video
Ok, the above video is an .asf file. The web cam I was using melted during the test and the .asf cache file was all I could salvage since the laptop locked up when the camera melted. Windows Media Player should be able to play it (mine does anyway). If anyone can convert that to a .wmv for me, I'd appreciate it. I wasn't able to find any free software that would convert it. As it is, you cannot fast forward it, so please excuse the long trip down my driveway.
At any rate, you can see from the video that the TC actuator is sticking. After the initial 2nd gear run, it doesn't release at 3000. Actually, it doesn't release until I've had the clutch in for several seconds and have stopped (you can hear my hawk pads squeaking just a little as the car stops). It should return to its initial position at 3000 rpm, and from the video you can see it doesn't usually do that. A good pressure side Turbo Control solenoid fixed it.