3,000 RPM Hesitation Fix

Last updated: May 3, 2003

There are several potential causes of and fixes for the hesitation that some people have experienced around 3,000 RPM:

• 3K Hesitation Survey Results (Read This First!)
• Grounding
• Battery
• Fuel Pump / Pressure
• Loose Turbo Vacuum Hose
• Double Throttle Control System
• Secondary Injector Solenoid Resistor
• TPS
• O2 Sensor
• MAP Sensor Filter
• ECU (Engine Mgmt Computer)

3K Hesitation Survey Results

See below for the original survey form that was mailed out.

CONCLUSIONS:

• The majority of the cars did not originally have this problem, but it developed over time.
  
  
• Some people reported that the problem only happened when the car was cold. Since this question was not originally asked, it is significant that that many people would have volunteered that information.
  
  
• Modded cars seem to have the problem worse than non-modded ones.
  
  
• The 94s seemed to usually develop it a after a lot of miles (30K+).
  
  
• Only two people tried having the injectors cleaned. This worked for one of them. Since the injectors are involved in the transition, it would be interesting to see more people try getting them cleaned.
  
  
• Reprogrammed stock ECUs did not seem to fix the problem, but some did. No one vendor had uniformly good results. The Apexi Power FC did seem to always work, but it is more involved and more expensive. The PFS PMC fixed the hesitation for some people, but not others.
  
  
• Many people mentioned sacrifices of chicken entrails, chanting, the invocation of spirits, and high colonics :-)

PROPOSED METHODOLOGY TO FIX:

1. Send the injectors out for cleaning. They probably need it anyway. Several old posts show that the injectors are flowing poorly at around 45K miles or even earlier. Cost: around $120 if you remove them yourself. This might take some time to remove them, and you will definately have down time while they are shipped to RC or Marren.
   
   
2. Clean or add more grounds. Even though this survey showed that it did not fix the problem for very many people, it is low cost, and not much time is required to do it.
   
   
3. Test or just replace the battery. Most people report that the stock ones are junky anyway and last less than 3 years, and it should cost less than $100 for a decent stock-type battery. Or look on this as a good opportunity to move up to the N-Tech BMK or to relocate the battery to the bins. Again, it did not work for that many people, but in case it would fix it for you, it is an easy thing to try.
   
   
4. Replace the plugs and wires. This didn't seem to help a single person in the survey, but was mentioned in previous posts. Plus your car is probably due for it anyway. The Mazda recommended interval is 30 months or 3 years, but should really be done more often.
   
   
5. Check the Double Throttle System. This only helped one of the 7 people who tried it, but it is low cost if you do it yourself and is related to the warm-up of the car. A lot of people mentioned that they only had the problem when cold. Not all of the people who mentioned the cold problem had checked this system.
   
   
6. Check the charge relief valve hose. Free to look at it.
   
   
7. If none of the above work, or if you have the money, replace the ECU with a standalone unit like the Apexi Power FC, Haltech, Electromotive, or Motec.

RAW DATA:

Number of reponses:			46

Number of 1992 cars:			 2
Number of 1992 cars w/ hesitation:	 2

Number of 1993 cars:			22
Number of 1993 cars w/ hesitation:	15

Number of 1994 cars:			19
Number of 1994 cars w/ hesitation:	10

Number of 1995 cars:			 3
Number of 1995 cars w/ hesitation:	 2


Number of US cars:			43
Number of Canada cars:			 2
Number of Singapore cars:		 1


Number that tried:  
    Grounding:				30
    Fixed it:	 			 4

    Plugs & Wires Replacement:		29
    Fixed it:			 	 0

    Battery Replacement:		24
    Fixed it:			 	 3

    Injector Cleaning:			 2
    Fixed it:				 1

    O2 Sensor Replacement:		14
    Fixed it:			 	 0

    Double Throttle Solenoid...:	 7
    Fixed it:				 1

    Charge Relief Valve Hose:		 7
    Fixed it: 				 0


Number of modded cars:				 9
Number of Modded cars w/ hesitation:		 5

Number of non-modded cars:			37
Number of non-modded cars w/ hesitation:	24


Mileage at which hesitation started:  

         0 - 10,000:	6 cars
    10,001 - 20,000:	0
    20,001 - 30,000:	6
    30,000 - 40,000:	5
    40,001 and up:	6

Mileage of cars w/ no hesitation:  

         0 - 10,000:	1 cars
    10,001 - 20,000:	1
    20,001 - 30,000:	1
    30,000 - 40,000:	2
    40,001 and up:	3


Number reporting it only occured when cold*:  8

NOTES ON THE SURVEY:

• The mileage from cars whose owners said it was there when they bought it used was discarded (except when the cars were purchased with under 10K miles) . (No way of telling when it started.)
  
  
• * = The number of people reporting the problem only happened when cold was not on the original survey, but a lot of people volunteered this info.
  
  
• When a fix worked for more than 10,000 miles, it was counted as "fixed", even if the problem returned later (could need the fix repeated).
  
  
• There was way too much data to try to correlate things like "replacing the battery worked for 94s but not 93s".

ORIGINAL SURVEY FORM

Year of Car:  ______ (92.5, 93, 94, 95, etc.)


Country:  ______________ (e.g.- US, Japan, Europe, Oz)


Mileage at which hesitation started to occur:  _______________


Mods at that time:  _________________________


Common Fixes Tried:  
    Grounding: 			_______  (Y/N)
    Plugs & Wires Replaced:	_______  (Y/N)
    Battery Replaced:		_______  (Y/N)
    Fuel Injectors Cleaned*:	_______  (Y/N)
    O2 Sensor Replaced:		_______  (Y/N)
    Double Throttle Control Solenoid 
      Check Valve & Resistor:	_______  (Y/N)
    Charge Relief Valve Hose:	_______  (Y/N)


Other Fixes Tried:  _________________________


* By cleaning fuel injectors, I mean sending them out, not just dumping a can of cleaner into the gas tank.  

Grounding

Date: Thu, 04 Sep 1997 08:51:02 -0700
From: derek_vanditmars@dynapro.com (Derek Vanditmars - RD)

Since my original posting of to supplement Mazda's grounding there has been allot of activity on this topic. I have been reading an noting all of the symptoms, results, methods that worked, and additional info. Many thanks to all of the people who have experimented on their cars for the benefit of the rest of us.

Here is a summary of this, enjoy:

SYMPTOMS & CONFIRMED RESULTS:
 ---also known as the ~3K RPM Problem
 ---hiccups anywhere from 2700 to 3300 rpm.  Sometimes it's 
	a very brief lean-out, other times it lasts seconds.
 ---when driving with lots of electrical accessories on the 
	situation was worse.
 ---my 3K rpm problem is MUCH reduced if not gone
 ---low-load rpm points where all the problems exist.  No 
	lurching, VERY MINOR hesitation.
 ---idle is smoother
 ---when trying to hold the car at a constant speed on a level 
	road where there is little load on the engine it would 
	buck (in a whiplash sort of way) kind of like it was 
	going to die but never did. It would not do this when 
	accelerating or when decelerating. It was really quite 
	annoying. You either had to be on the gas (accelerating) 
	or off the gas.
 ---a broad 2800 to 3500 flat spot
 ---3K rpm problem is augmented by higher electrical loads 
	(i.e., a/c, headlights, etc.)
 ---Magic.  Just Do It.
 ---The car runs a whole lot smoother.  I do not get a jerk 
	when I hit the throttle and I always used to.  My 
	shifts are smoother with the more consistent throttle 
	response.
 ---I put on the ground strap and now I can't go above half 
	volume.  So, all you people out there complaining 
	about the low volume from the radio: the radio is 
	probably fine and the ground isn't.  I wonder if 
	my head lights are brighter now.


WHAT WORKS:
 ---Check and clean ALL ground connections, this includes 
	removing any oil/crud, re-torque connection, and 
	sealing with a spray-on battery terminal coating.  
	Here is a list of ground points to check:
        --Battery terminal
        --Main engine ground (to the rear of sparkplugs 
		below oil filter)
        --Engine ground (connects between intake and 
		firewall)
        --Exhaust ground (after CAT)
        --Any additional grounds that have been added
 ---Add a wire from the negative terminal of the battery 
	to the car chassis.  Anything from a 10 gauge to 
	a battery cable size seems to work, (I recommend 
	a battery cable size).
	Chassis points used:
        --driver's side wheel well using one of the bolts 
		that attaches the most front bracket with 
		fuse/relays/etc..
        --to a bolt holding the windshield washer reservoir 
		on the inner fender.  The problem was I had 
		placed the cable end on top of the plastic 
		and run the bolt through the cable-end.  It 
		may have had some grounding affect, but minimal, 
		I'm sure.  I moved it to a blank threaded hole 
		that is a little ways back on the fender well.  
		Good contact now.
        --Added a ground wire from the battery to the closest 
		bolt which holds the front fascia to the frame 
		(in front of the A/C bottle),
        --Use 10 gauge insulated (multi-strand) wire, and a couple 
		of crimp on connectors with big enough holes ("ring 
		terminals") to accommodate the factory body bolts 
		and the bolt through the negative cable clamp.  Any 
		bolt which screws into the frame will do.  I'd 
		prefer to use one that didn't have plastic on one 
		side which is why I avoided the washer bottle 
		mounting screw.
        --clamped one with the ground cable to the battery and the 
		other to the unpainted aluminum tray at the front.

ADDITIONAL INFO:
 ---To ensure a long lasting ground connection it needs to start 
	out clean, establish a metal to metal contact, and also 
	needs to be protected.  Start out clean by scraping any 
	paint and removing any oil or dirt.  To establish a metal 
	to metal contact use external tooth washer between the 
	lug and the metal you are connecting to, (outside 
	diameter = lug diameter) and torque the sucker down.  
	For protection of the ground connection and any bare 
	metal use a spray-on battery terminal coating, most 
	automotive stores sell this stuff.
 ---Several people noted that the Main engine ground (to the 
	rear of sparkplugs below oil filter) on their car was 
	loose or getting very hot, so this connection is a 
	MUST DO for checking and cleaning.
 ---Yeah I already have the additional ground on my battery cable, 
	so make sure it is still doing a good job.
 ---Battery cables can be purchased at a local NAPA or similar 
	automotive outlet.
 ---The engine mounts are totally rubber/plastic and they do not 
	conduct electricity
 ---The battery, engine, and the chassis must be connected together 
	with a good ground.  Do not worry about ground loops, as 
	all you need to do is establish a firm connection to the 
	car chassis for the battery and the engine.
 ---Checked out some later model automobiles that have heavily 
	computerized engines, (piston type ughh!) and there are 
	several ground connections, (big fat flexible straps) 
	between the engine and the car chassis/body.  In 
	addition they all had a big fat battery cable between 
	the battery and the car chassis/body.  Not too sure what 
	Mazda was thinking with the factory ground set-up.

Hope this helps you all, eh!

____________

Date: Fri, 22 Aug 1997 13:33:57 -0700
From: derek_vanditmars@dynapro.com (Derek Vanditmars - RD)

The car ('93 Rx7) was running rough between 2,500 to 3,000 RPM, hesitating etc.....

I noticed that when driving with lots of electrical accessories on the situation was worse.

Took the car to Mazda dealer and they added an extra ground connection on the battery's negative lead. They did this by removing the insulation and crimping on a lug (big) on the existing battery cable. The lug was then attached to the driver's side wheel well using one of the bolts that attaches the most front bracket with fuse/relays/etc.. They left the rest of the cable so that it still connected to the engine.

Big difference with this! In addition I checked out ALL other grounding points between the engine and the car chassis, ( the place where the battery negative cable normally attaches to the engine was loose and there was lots of oil/crud between the cable lug and the engine). Removed and cleaned and put back together. Now with all of the electrical stuff on, no noticeable hesitation and runs as smooth as a rotary should.

Hope this helps, eh!

____________

Date: Sun, 24 Aug 1997 00:30:34 -0400 From: Daegal Benedetto

I added the extra ground wire to the locations you described and guess what? No more 3K hesitation, MY MAN!! I had a feeling it had to be something simple like this,and it actually makes a lot sense . If it were a problem with the programming of the ECU everyone's car would act the same, which is not the case. I don't have a problem with appliances loading down my system, but I would like to check he motor ground as well. Could you please save me some time and tell me where its located. Thanks again, You Da Man!

________________

Date: Sun, 24 Aug 1997 21:23:29 -0400
From: Daegal Benedetto

I used a 1'-1'1/2 length of 12 gauge wire and crimped spades with holes onto the ends. I unscrewed the nut on the neg. terminal clamp, brushed the side of the clamp with a wire brush and put the spade over the bolt and put the nut back on.

On the other end, I took off the relay/fuse box on the drivers side wheel well ( nut on one side, snap on the other)to get to the bolt directly under it. No need to undo any wires, just push it over towards the front. I removed the bolt and sanded off the paint around the hole(I don't know if that's necessary) and attached the other end of the wire. While I was there I took off the factory ground(one nut) and cleaned it up with the wire brush.

Total time: 20 min.. Total cost, pocket change. I'm a proffesional audio engineer and have lots of 12 gauge with tons and tons of strands that we use for grounding recording studio components. The more strands the better, as electrons travel on the outside surface of the strands. Simply, more strands=less resistance. This type of wire can be found at hobby shops that sell RC cars. A company named Trinity sells it but the best is is Draggin' Wire from a company called Stage III. Both of these come with high heat silicone casing. If you can't find any, let me know and I'll send you a length of the wire I have, good luck.

_____________

Date: Fri, 29 Aug 1997 15:12:37 -0600
From: Whitteron@sisna.com (Chris Whitteron)

My hesiation problem seems to have gone away. It's one were all though it can occur at any time, it seems as though it is augmented by higher electrical loads (ie, a/c, headlights, etc.) So here's my experience:

First off let me say that I was premature in saying the 60,000 mi. tune-up seemed to have fixed my problem last weekend. The hesitation came back the next day with a vegenance after 60 miles. So I decided to try grounding out the battery with an additional wire as described by Derek.

I used an approx. 18" length of 10 ga. wire with an o-ring connector on one end and nothing on the other end. Connected the end with the o-ring to the bolt that holds down the washer fluid filler and the other end to the battery.

I have been driving in this configuration since Monday (Friday afternoon now), for around 200 miles, without any noticable hesitation and smoother throttle response(doesn't kick as hard when the accelerator is initially depressed). Though I questioned whether this was the solution since I had to disconnect the negative terminal, and a few people have been saying that this resets the ECU over the past few days.

So this morning, I disconnected the o-ring end and taped it off, ensuring that I didn't disconnect the entire negative terminal and drove it, immediately the throttle response deteriorated and it was hesitating violently shortly after. After 40 miles I decided to reattach the additional ground wire, once again NOT removing the negative terminal, and instantly it was smooth again. Drove it for 15 miles, no appreciable amount of hesitation, and throttle responce was much better. I am convinced that this was my solution.

BTW, over the past week I have changed nothing on my car, except this one ground wire. Furthermore, ALL driving done this morning was with the A/C on and at 2, as well as the radio on, to increase the electrical load.

Hope this helps everyone too, it seems to have fixed my problem.

______________

Date: Thu, 14 Oct 1999 12:45:48 +0100
From: gpc@amlibs.co.uk

>My car had quite bad bucking between 2500 and 3000 RPM under light load. An
>improved manifold ground strap solved the problem, ie not a single
>recurrence over 2 months.

I don't know if this is related or not but I thought I'd mention it on the list just in case it helps someone. ...reminded me of something I found after fitting an additional ground strap to the throttle body. Before fitting the strap I found that when checking the voltages on the TPS both wires, 3F and 3G would go way above the normal maximum voltage when rotating the throttle linkage. This was even though the idle voltage was well within spec.

I initially thought the TPS was bad and bought a new one.... But this one was exactly the same!

After fitting the additional ground strap some months later and then at a later date checking the TPS again I found the voltages were within spec at all times.... Could just be coincidence but I reckon the better grounding contributed to this. I must admit though - I've never had the 3K hesitation problem (although I did have a problem with bucking when coming on/off throttle)! The FD in question isn't US spec either - its a Japanese import.

______________

There were more posts almost all saying that this helped if not entirely cured the problem. I do not have this problem, and my car is almost new (3,700 miles), so maybe the ground wire has not worked loose or built up crud on it yet. This sounds like the magic bullet. Other possible causes are included below. --Steve.

______________

Date: Sun, 14 Sep 1997 22:25:16 -0700
From: derek_vanditmars@dynapro.com (Derek Vanditmars - RD)

In addition to adding the extra grounding wire from the battery to the chassis of the car, you MUST also check the ground cable to the engine block.

Follow the factory ground wire from the battery, and you will end up behind the spark-plugs and below the oil filter, (kind of way down there). The wire has a ring temrinal lug that is attached to the engine with a bolt. This should be removed and cleaned, (ring lug, bolt, and engine block) then re-attached.

To establish good grounding, the battery AND the engine block need to be connected very well to the car chassis.

Since the 3rd gen RX-7 seems to be sensitive to grounding, and with aftermarket stereo systems there is a large additional load put on the electrical system. With the additional amps required, you may want to add another ground wire from the engine block to the car chassis. This would be in addition to the extra one on the battery.

______________

Date: Mon, 18 May 98 10:41:48 EDT
From: zgluszek@VNET.IBM.COM

There are two approaches you can take:

1). Regardless how good the ground cable looks, measure DC voltage drop from
     the negative of battery to the ground terminating point:
     A).  Engine block.
     B).  Frame (This is the one which suppose to fix 3K hesitation.
     B).  CPU reference ground.

2). Instead of DC voltage drop measurement, go to hardware store and buy
     10 gauge multistrand wire.  The wire usually cost less then 15C / foot....
     Use it to run temporary (test) ground connections from:
     A). Negative of battery to engine block.
     B). Negative of battery to frame.
     C). Negative of battery to CPU ground terminal.  This one you can
    temporary route on the ouside of the car, just run it from
    underneath the hood through the door opening. Engine bay and
    door gaskets will compress.  If this ground fixes the problem,
    then you look for a more permanent solution by running it
    properly through the fire wall.....

If there is an improvement then you can check which temporary ground fixed it, by removing them one by one and checking car response. This procedure proves 100% if you are running a ground loop problem. Just remember to go to negative of the battery with each run.

I personally prefer the second method, because in the first method all you can measure is the DC component between the reference points. If you are running into a high frequency noise problem, this will not test it. you would need to use oscyloscope which is not to feasible while running the car. This is generic procedure I used when troubleshooting my BMW....

Second possibility is due to the secondary injectors coming on (but I think that they would come on closer to 4.5K not 3K). If the fuel filter is clogged the transition to 4 injectors would cause temporary higher then usual fuel pressure across the fuel rail causing temporary lean condition. Also you wouldbe loosing power at higher RPMs.... Per your description - this does not sound like the case....

Another area I would check are the power and ground leads to fuel pump. If there is a voltage drop you are not running at the full pump output. Remember that the pump is fed through the resistor at low RPMs and once there is a higher fuel demand, a relay controlled by CPU bypasses the resistor supplying the pump with the full battery voltage. If there is a corrosion on power / groud feeds leading to fuel pump, causing voltage drop across the contacts, the pump output would be limited. This one requires checking voltage across the pump connector on the top of the gas tank, which is major pain in the A$$!

Then there is a possibility that the fuel pressure regulator is not holding proper pressure and when the fuel pump goes to high output it drops the pressure to much before it stabilizes. This could be tested with fuel pressure gauge.

I am assuming that your ignition system is functioning properly, that the spark plugs are in good condition and there are no engine sensors outside of their operating range....

______________

Date: Tue, 16 Nov 1999 12:31:38 EST
From: LikuidG@aol.com

Well, these 3k hesitation posts are getting really old so heres a fix that I did on a couple of Chicagoland area RX7's

1. Put thick extra ground strap from extension manifold to firewall (use multi-strand stereo cable, and clean metal using strong wire brush)

2. Clean the ground lug at the bottom of the motor and move it to a different opening on the housing

3. Replace stock cat back ground with multi strand stereo cable

4. Affix a new ground point from downpipe/precat to frame (use multi-strand stereo cable, and clean metal using strong wire brush) the key here is to put the ground strap on the downpipes flange, by putting it on the main cat flange electrons are somewhat restricted by the gasket. At this point the O2 sensor is grounded suffice.

5. Replace battery (if needed)

6. Clean fusible links. Right next to your batterys' positive cable theres a little fusible link box there (you should see a "Main 120" on it, thats the box im talking about) you need to take off the cover and you should see the positive and negative in and outs clean the contacts thoroughly with a wire brush. This is a very important step

7. MOST IMPORTANT STEP!! Make sure your Alternator belt is tension to stock specs. This is the main cause of the 3k hesitation. I would suggest buying a new belt and make sure it's tightened to specs. Please don't underdrive this pulley, this is a very important pulley in our car.

8. Reset your ecu by disconnecting it and pressing the brake for 5 seconds

This should work, I did this on several 93-94 RX7's. People out there think the 3k hesitation is a voltage problem but its an AMPERAGE problem. the ECU needs to see a consistent amount of AMPS especially during the point the secondariness come on. The manufacturers of the 93-95 RX7s alternators (Mitsubishi?) used cheap parts and I come to believe that the 2nd gen alternators, although they are basically the same alternator were manufactured better back then.

______________

And you read it here first.... --Steve:

Date: Wed, 25 Mar 1998 22:29:11 -0500
From: flyinbeans@juno.com (Bryan)

This is for anyone with the infamous hesitation problem.

     Bulletin No.:	005/97
     Issued:		3/10/97
     Applicable Models:	1993-1995 RX7
     Description:		A jerk or bucking condition may occur under any of the
				following conditions:
				- Under light to moderate acceleration
				- Cruising at engine speed of 2000-2200 rpm
				- A/C on

"This concern may be improperly grounded engine harness, creating high resistance in the engine ground circuits. This condition affects the fuel control system.

Customers complaining of this concern should have the vehicle inspected and if necessary, repaired according to this bulletin."

The rest of the bulletin outlines a detailed repair procedure of cleaning and retorqueing ground straps. Specifically it calls out the wiring harness ground, main battery ground. Also, it calls for the replacement of the ground strap (earth wire).

A hardcopy of the entire TSB with detailed repair instructions and illustrations is available FREE (*they sent me mine for free..I assume they would do the same for anybody*) from Technical Information Services (TIS) which is a branch of the NHTSA. Basically, it involves faxing a request to TIS. If you go the NHTSA's web site, you will find the 800# for TIS. If you call that number, a recorded message will tell you exactly how to request this info. Hope this helps others, cause it sure helped me.

________________

Date: Fri, 15 May 1998 17:03:40 -0700
From: Mark An

Someone mentioned to the list a while back ago on how to receive TSBs from the NHTSA. I have followed the instructions and the results are good and bad.

The good part is I received the detailed TSBs I requested from them. Very detailed TSB's with pictures of replacement parts and pictures of the replacement procedures.

The bad part. No one mentioned they would invoice you for their effort in retrieving the TSB's for you. I got invoiced for $20.50 for copies of 2 TSB's. Geez!

One of the TSB's I got was pretty interesting. This will interest those with the ~3k hesitation. The following is from the actual TSB:

Applicable Models: 1993-1995 RX-7

Description: A jerk or bucking condition may occur under any of the following conditions:

    - Under light to moderate acceleration
    - Cruising at engine speed of 2000 to 2200 rpm
    - A/C on

This concern may be caused by improperly grounded engine harness, creating high resistance in the engine ground circuits. This condition affects the fuel control system. Customers complaining of this concern should have the vehicle inspected and if necessary, repaired according to this bulletin.

Repair procedure:

1. Clean and retorque harness ground.

        a. locate ground on left side of engine, mounted to bracket
           behind A/C compressor.  Refer to the applicable BETM or wiring
           diagram.
        b. Tightening Torque: 69.5 - 95.4 in-lbs

2. Clean and retorque main battery ground and bracket mounting bolts
       (bracket used for mounting the ground).

        a. Tightening Torque:    69.5 - 95.4 in-lbs
        b. Remove the original ground strap and terminal bracket between
           the engine hanger and the bulkhead.
        c. Install new style ground strap using the original bolts.
        d. Tighten Torque: 12 - 17 ft-lbs (engine hanger side)
           69.5 - 95.4 in-lbs (engine room bulk-head side)

3. Verify Repair.

Parts Information:

    New Part#: FD02-67-E70
    Old Part#:  FD01-67-E70A

Warranty Information:

Warranty type:	A
Symptom code:	08
Damage code:	9S
Part# Main Cause:	FD02-67-E70
Quantity:		1
Op. #:		XX899XRX
Labor hours:	.2 hrs.

________________

Date: Sun, 19 Apr 1998 08:54:16 -0600
From: David Beale

The engine grounding wire may have to carry high current. Remember that the alternator is on the engine. The ground path from the alternator often goes through the engine and engine ground strap(s). I've seen this problem on several cars (one MELTED the ground strap, the other only melted the nylon housing of the ground strap connector - a TII).

A good test would be to put an accurate voltmeter on the battery and, with the engine running at a reasonable RPM, say 3000, note the voltage. Then turn on the lights, rear window defogger, stereo, etc. and note the voltage again. If there is a difference, measure between the engine and the chassis with the voltmeter - the missing voltage should show up there if the strap is the problem. These "ground problems" can be very tricky to diagnose.

Battery

>>This brings up another point I would like to mention.  I have heard some
>>people mention that a new battery made the stumble less apparent or even
>>made it disappear.  This makes absolutely no sense to me.  The ONLY
>>reason for a car battery is store energy for starting.  I'm suprised no 
>>one has agreed!

> I have always believed that battery also smoothed out the electrical signal
> from the alternator.  An alternator puts out "pulsed DC" after going through 
> a rectifier...  Right???

Fuel Pump / Pressure

    >If that is the problem all you need is a rpm switch to apply 12v at say 2800 
    >just enough to bet the ECU to the punch. Like a msd switch or such. 

Double Throttle Control System

	" The response delay of the pressure sensor followed mounted 
	by rapid acceleration temporarily causes a lean fuel mixture.  
	The double throttle control system prevents hesitation caused 
	by this lean fuel mixture by slightly delaying the opening of 
	the double throttle valve after the secondary throttle valve. 
	The double throttle valve is controlled by the ECU through the 
	solenoid valve."  

>DOUBLE THROTTLE CONTROL SYSTEM DESCRIPTION
>
>The response delay of the pressure sensor followed mounted by rapid
>acceleration temporarily causes a lean fuel mixture. The double throttle
>control system prevents hesitation caused by this lean fuel mixture by
>slightly delaying the opening of the double throttle valve after the
>secondary throttle valve.
>
>The double throttle valve is controlled by the ECU through the solenoid
>valve.
>
>(diagram showing that the solenoid is activated (thus closing the double
>throttle) by the conditions labeled "IDLE ZONE" or "COOLANT TEMP BELOW
>80C [176F] | START-UP ACCELERATION (AT)")
>
>OPERATION
>
>When one or more of the above conditions are met, the ECU turns the
>solenoid valve ON, applies vacuum to the actuator (double throttle
>control), and closes the double throttle valve.

Secondary Injector Solenoid Resistor

Loose Turbo Vacuum Hose

Date: Wed, 22 Oct 1997 02:07:51 -0400
From: Daegal Benedetto

Jake Watkins wrote:

>> If it's a 93 you should be sure to replace the "U" shaped
>> vacuum hose for the turbo relief valve (I think). This made
>> a difference on mine even before I did the ground fix.

HESITATION AT 3000 RPM

Not a TSB. Published in January 1994 Mazda Tips (Dealer newsletter)

A kink in the turbo charger relief valve vacuum hose may also cause the engine to lack power or hesitate during acceleration. Check the hose and replace it with a modified one if it is kinked. The new part is available at the PDC (part # N3A3-20-341A).

The part was fixed from 94 on, but I'd check it. --Dae

______________

Date: Tue, 21 Oct 1997 23:59:12 -0700
From: "Hung-Jen Hung"

This is the part dealer replaced for me to fix my 3k hesitation in addition to the newer ECU w/ a ground wire. After replacing this part, my 3k hesitation was gone, but after several months it came back. :( Probably it's time for me to replace it again.

TPS

From: Joe Ramos (jramos@sunup.corp.sun.com)

Richard Thomason reminds me that perhaps it's something with my TPS. Anyone worked/adjusted this thing who can give me some tips. As I recall (sitting here in my office without my shop manual) it's at the rear of the engine, a real pain to get at. Would it be worthwhile just to install a new one? If so, what do they cost? Joe

___________

Date: Mon, 25 Aug 97 07:01:20 -0500
From: "Linthicum, Sandy"

Cost for a new TPS from Mazda Comp about $80. Its right next to the hose under the intake that goes to the factory boost sensor. There is about 2 inches of working room between the TPS and the firewall. It can be done, my mech replace my TPS about a month ago with out removing the intake. The two screws holding it on are unfortunately not hex head, but phillips, and in my case were very tight. Main problem is getting the screws loose.

I replaced the TPS because of light throttle surging but it didn't help. Turned out the O2 sensor was bad.

O2 Sensor

Date: Fri, 1 Aug 1997 12:21:46 -0500
From: "Kevin T. Wyum"

- -----Original Message-----
From: Dana Bourgeois [SMTP:fg@portal.com]
Sent: Thursday, July 31, 1997 10:36 PM

I'd like to point out that someone (sorry, I didn't catch the name but you know who you are) with a non-sequential setup just recently posted that he also has the hesitation, it went away with an unplugged 02 sensor but it came back later. I think that this means you can narrow your search to only those components left after a non-sequential conversion.

(I think the above was Dana's post, the rest is probably Kevin's reply? -- Steve)

Most of us have known for quite a while what is causing the bucking and hesitation around 3000rpm in the thrid gens. It's no great mystery. The problem that everyone is having is trying to correct it. If you scan through the shop manual you will see that only one thing really happens at this point, the secondary injectors come on. It's not strictly based on RPM, but instead load condition, which is why it doesn't always happen at the same RPM. Everyone should have figured out by now that you can drive around the problem by lowering the load, ie: go to a lower gear. The bucking is happening when the car is on the break point of the load map. They keep turning on and off, which is why if you are hard on the throttle you don't notice any bucking because it is a smooth transition through that cycle and there is no on off on off...

The solution that was tried with the PFS computer was to make it super rich throughout that RPM range. The problem with this is as I mentioned it is a load point not an RPM point. It is like using a shotgun with buckshot to kill a fly buzzing around. Additionally you are not controlling whether or not the injector turns on, only increasing the total percentage of fuel from whatever happens to be on. If you have datalogging you will see it happen. The RPM etc are going up and the pulsewidth as well and then suddenly the pulsewidth nearly cuts in half. This is the secondary injectors coming on. Anyway if somebody were trying to solve this they would have to come up with a way to keep the ECU from returning to the loadpoint just before the injectors turn on. Which is probably not very feasable and would cause a number of collateral problems. The other option would be to extend the full pulsewidth of the primary injectors, or create more of an overlap of the two. The most effective way around the whole problem though is to modify your driving style. Don't drive in that load range, use a different gear etc.

The O2 sensor things as far as I can see is a simple way to richen everything up due to the error with no feedback. It probably defaults to a overly rich safe mode.

Kevin T.Wyum

______________

Date: Mon, 16 Mar 1998 22:42:08 -0500
From: Tuck

(If) the hesitation only occurs under acceleration, I don't see how the O2 sensor could possibly be the culprit, as the computer only uses the O2 sensor for feedback during closed loop conditions. The hesitation, from what I have been able to discren, is simply the result of poor programming on mazda's part.

______________

Date: Wed, 06 Aug 1997 11:41:28 -1000

From: richardt@lava.net (Richard H Thomason)

After finally getting my car back from the bodyshop, I decided to expiriment with disconnecting my heated 02 sensor. My car has always had the 3k hesitation to some degree or another. Adjusting the TPS and replacing the old 02 sensor helped, but especially after pulling the cats the car got a little "surgey" at part throttle and would sometimes buck like a bronko for no apparant reason. Switching to Pettit's upgraded stock computer also helped for about 2 weeks, but the conditions returned gradually. Keeping new plugs in the car also helps, but the car was never what I'd call smooth except at WOT.

BTW, my a/f meter clearly shows the momentary part throttle lean condition at 3k described by others, but does not show anything weird to explain the other surgyness.

Like Kevin, I have simply altered my driving style to avoid the "burp" at 3k, but have always been irritated by the other surges at low rpm, which I assume are caused by a little bit too lean mixture(even though my meter does not confirm this).

Anyway, after disconnecting the wire from the 02 leading to the computer but leaving the other wires in place (so my a/f meter still works and the 02 is still heated and grounded) I found that SO FAR both the 3k hesitation and the other surges are gone. My meter shows the car running about 13.1 at steady state part throttle rather than oscillating around 14.5 and lower (leaner). Full throttle is of course the same.

After about 4 days of driving, the check engine light finally appeared, but there has been no change in behavior. Perhaps the computer will eventually tailor back in all the part throttle grumpyness or worse yet, make the part throttle operation excessivly rich, but so far I really like the way the car is driving. I tried to pay particular attention to tip in throttle response in 5th on the freeway and feel that if anything, the car feels a bit more responsive when I hit the throttle at 2/3k.

The only downside so far is that gas mileage must surely be worse with the slightly richer conditions (don't know, don't care). Also, my EXPENSIVE 10.5 plugs may yet crap out on me even sooner than before (do care).

Anyway, as it stands now, the car is smoother than it ever has been since stock and still goes as fast as ever at full throttle. Unless something changes, that wire is staying off.

_____________

Date: Thu, 18 Sep 1997 19:04:48 -0500
From: David Liberman (liberman@scott.net)
Subject: Re: (rx7) [3]oxygen sensor replacement

The oxy sensor from an '86 Ford Escort will work fine; it's a three-wire sensor, too, so if you plan to run leaded race gas it's a necessity. If not, then don't wire it to 12v, just hook up the sensor wire. And, it's $38.99 at your local NAPA.

________________

Date: Mon, 03 Nov 97 07:07:27 -0500
From: "Linthicum, Sandy"

(The 02 sensor is) NAPA part No. OS204

Should have 3 wires (2 white, and 1 black) hook the white wires to pos 12, and gnd, and hook the black wire to the wire where the old one hooked up.

MAP Sensor Filter

Date: Tue, 28 Mar 2000 12:48:56 -0800
From: "Ulen, Robert S" (Robert.Ulen@PSS.Boeing.com)

It's possible that a semi-clogged MAP sensor filter could contribute to the 3000 rpm hesitation syndrome. IMO, the 3K hesitation is probably a whole bunch of factors, and as each factor varies somewhat on different cars, each car will show a different level of hesitation. It would be *very* interesting for someone who has a bad case of the 3K hesitation to put a new MAP sensor filter on to see what effect it has.

>>>It's definitely a filter to keep any oil mist out of the MAP
>>>sensor.  Remember the MAP senses absolute pressure,
>>>which means air is flowing both ways through the tube
>>>connected to it when going back and forth between vacuum
>>>and boost pressure.
>>   
>>Theoretically, if the filter was semi-clogged from a lot of turbo
>>oil mist, the MAP sensor could become "sluggish" in its
>>response, and may cause running problems under highly
>>dynamic situations (quick engine rmp changes).

_______________

Date: Tue, 28 Mar 2000 21:46:40 -0800
From: "Derek Vanditmars" (dvandit@istar.ca)

I bought a new one recently an it made no eFing difference at all. I was thinking the same kind of stuff, like maybe it is delaying what the boost sensor sees.

If you want to check this out just run without the filter and see what happens. I would also highly recommend that you put it back asap.

_______________

Date: Tue, 28 Mar 2000 22:10:51 -0800
From: "Derek Vanditmars" (dvandit@istar.ca)

From what I have noticed, the 3K hesitation occurs under "light-medium load" conditions and "partial" throttle, the manifold pressure is fairly stable during this event, so I would think that a sluggish MAP would not be the problem.

ECU (Computer)

>>I have 8K miles on the car and do not have the 3K hesitation.  If this is something 
>>that happens gradually over time, my guess it is not the ECU.  The ECU
>>programming should not change.  
>
>Don't forget that theres a 20K miles 'switch' built into the ECU.  I couldn't
>say with any certainty what exactly this switch does but I guess its to help
>with running in a new motor.  The change in the way the ECU controls the
>fueling/timing/boost/etc could be why you haven't noticed it yet...

[Mail me] [To Lightning home] [To my home page] [Copyright Notice]