Akagis_white_comet

iTrader: (4)

I'm officially at my wits end and sick of ******* around with starters.

Car is 1987 GXL with 20B-REW, T2 trans/flywheel/clutch & starter. Battery is brand new Deka Intimidator 9A48, relocated to passenger storage bin with 4awg welding cable for all connections (soldered lugs). Each connection wire wheeled to bare metal, given dielectric grease and torqued to 10ft.lbs per the FSM. All cables, car body, underhood bus bars and 150A breakers have 0.7 ohms resistance.

My starter was acquired from a fellow forum member on trade for a N326 ECU over a year ago and won't spin past 175rpm on the engine. I believe it spins at 2100rpm on the bench (done by audio frequency x 8.08, see my build thread for more info). After 10 seconds of cranking, it trips the 150A breaker.

A new starter from autozone did approx. 2882 on the bench and 70 rpm on the car, then tripped the breaker after 10 seconds of cranking. Let me repeat that...SEVENTY, as in 7-0 RPM

Let me confirm that it is NOT the battery, the wiring or the engine. This Deka Intimidator is brutally powerful and took 10 crank attempts before breaking a sweat. The old autozone battery keeled over after 4-5 attempts. Every cable throughout the car is 4AWG welding cable with proper lugs soldered on, with exception to one being an off-the-shelf 4awg battery cable with the lugs crimped on. Every single conducting device shows 0.7 ohms of resistance, including the body itself between the engine ground point and battery ground point (rear seat bracket bolt hole). Engine turns over easily with a 1/2" drive ratchet and 1 1/4" socket.

Diagnosis: both motors have insufficient torque/rpm and excessive current draw under engine load. Best guess is an out of true commutator and possibly worn brushes too.

Are there any rotary shops that rebuild starters to actually work correctly? I need one that can actually not turn to **** as soon as the merchant collects payment. I've seen two pass on the workbench but fail when bolted up to an engine, one of which failed bigger than Jeremy Clarkson.

Realistically, a starter rebuild needs to have the following done to ensure quality:
Commutator measured for runout, trued on a lathe, replace if it cannot be trued back to specs
brushes measured and replaced if beyond spec
Bench current draw & RPM test
Current draw & RPM test when mounted to a dummy engine

Can someone point me in the right direction?

jackhild59

iTrader: (7)

I had a starter that gradually went bad on my 88 GXL. This was the oem starter on a 325000 KM Canadian car. It had a long hard life. The starter developed a dead spot. Then 1 out of 10 times when the starter stopped in the dead spot, no startey.

I bought a reman starter from Orielly's. It wouldn't spin the motor fast enough to reliably start the car. It passed the bench test at Orielly, so I put it back on and checked all the connections, bought a new battery etc etc. It still wouldn't spin the engine fast enough.

Finally, I took it back. It still passed the bench test but I just insisted that it was bad.

They replaced it with another identical unit. That next starter worked GREAT.


Make Autozone give you another starter. Odds are you just got a bad one.

Molotovman

iTrader: (49)

Get a Bosch reman.

clokker

iTrader: (2)

I just replaced the brushes in my (presumably) original starter and that was all it took to wake her right up.
They're easily inspected/replaced (and relatively cheap - <$20 at Autozone) and are about the only component you can replace yourself (well, there's the solenoid...), so it's worth a try.

That fails, I guess you just keep replacing till you find a reman that works.

Akagis_white_comet

iTrader: (4)

Sorry Clokkar, but the point of this is to NOT keep guessing about why they come DOA and to have a logical reason about what is causing each one to not perform under load.

Aaron Cake has double-checked my cables and connections and they meet with his approval. By his reasoning, they are not the cause of the problem, not is the battery.

Starter brushes are $15 at autozone, $10.80 MSRP at the dealer. Oddly enough, you can only find them at autozone by putting in the Mazda part number (N326-18-X66). Commutator (AKA Armature) is the same between NA & turbo and shared with a bunch of cars. This is around $220 MSRP to replace, so I'd wager that the whole core starter is scrapped if the commutator is not easily reuseable. My guess is that the parts store rebuilt ones slap in new/used brushes if it turns slow or not at all, a new/used solenoid if it won't extend and ship it out.

I took the chinese special back to Autozone and they wanted to argue with me about how the mode of failure wouldn't show up on their bench tester. Gave them a crash course in physical and electrical engineering courtesy of Lynn Hanover and immediately got refund. This is why I have three policies regarding parts stores and dealers:

1. No part is ever pre-paid*.
2. It is not paid for untill it is proven to operate correctly per the FSM on my phone
3. No core will be returned untill the new item is installed and proven to operate in real-world conditions
*Ray Crowe at Malloy Mazda, Atkins Rotary, Banzai Racing, Mazdatrix and Rotary Resurrection are exempt from this policy, no exceptions for any reason on starters

Went across the street to Advance Auto and had them order the same part, insisting that I pay for it when it arrives. They wanted to not play ball, so I told them what happened with Autozone and they saw it my way. Their core charge is less, so it makes me slightly more happy. This is my way of exchanging it for another identical item, same part number at both places.

If this one is a POS too, then it's time to call Dan Atkins or Ray Crowe. Pretty sure both would be willing to test a starter under engine load before it is shipped out.

Leonel95

I had an issue just like yours. I assumed it was batteries too...

I had to even change the timing a bit to help the car start.

Mind you this was on a new aftermarket starter.
I went ahead and got a DEnso brand starter. Problem gone.

I remember some were on the forum that if the starter doesnt crank beyond like 250rpm and above a signal doesnt occur properly at the ecu. Hence it floods etc...

clokker

iTrader: (2)

Oh, I thought the idea was to get a starter working.
Sorry.

Akagis_white_comet

iTrader: (4)

ECU is Haltech PS2K. What you're referring to is the Max Cranking RPM, the rpm that determines whether the engine is running or trying to be started. Mine is set at 250 rpm per Claudio's instructions. This means that after 250rpm, the ECU pulls the cranking enrichment out of the fuel map.

Haven't ever seen a Denso brand starter for our cars, Leonel95. Got a link to one? You got me curious...

jackhild59

iTrader: (7)

I woulda been satisfied to just have the car start.


Sorry, I kept dozing off...I'll bet neither Dan Atkins nor Ray Crowe will be impressed with your personal policies and will not be willing to test a starter under engine load-*just* for you. Even if they were willing, do you really think they have a 20B sitting around for starter cranking load tests?

After all, it's a *starter*. You are really over-thinking this one.

Akagis_white_comet

iTrader: (4)

It's dead Jim. 70 rpm again.

10 seconds tripped the breaker and started to melt the cables purchased from Advance Auto. Pulled the cables to examine them. The heat shrink was loose and the terminals were burnt on the factory crimped ones or falling off on the ones I soldered correctly. Parts store battery cables are NOT the size they claim to be. These claimed to be 4AWG but are actually 6AWG and failed because they are Chinese ****.

Swapped them with 4AWG cables made a couple days ago from welding cable purchased at DeLille Oxygen and tested it again. 70 rpm and a tripped breaker after 10 seconds. These cables were not damaged because they were soldered correctly using rosin core solder and a butane torch.

Results: Starter is junk, cables are junk. Both were returned. As I had cut one of the cables in half to make into two cables (for ground), I'm stuck with it now.

Remy International makes both Worldwide and Duralast starters. Same part numbers, same price and the results were the same as well. Chinese/Mexican ****.

Going to visit DeLille again tomorrow for more welding cable & lugs, then reinstall the old starter.
Lesson: Don't buy chinese or mexican garbage. Get the good stuff!

AGreen

iTrader: (11)

If the places (china/mexico/kazhakstan) that rebuild these starters are re-winding the rotors of these things, they could be shifting the electrical/mechanical neutral planes. In order for an electric motor to make torque, the angle between the 2 magnetic fields has to be within a certain spec. A starter motor is a "series-wound" electric motor, which means the armature and stator field are in series with each other. They have good starting torque, but pretty poor speed regulation. A motor that spins at 2800 rpm sounds great, but that's probably because they've shifted the electrical neutral plane of the motor for better speed, but poor torque. In the end, with greater torque, you'll have a higher speed under load; conversely with a lower torque/higher top speed motor, it'll spin slowly under load. That also explains why it's blowing the breaker after 10 seconds. If the ENP has shifted too far, it's drawing a lot of extra torque due to "armature reaction".

Your best bet is to find a used ORIGINAL MAZDA starter, disassemble it, clean it, re-brush it, and try again.

j9fd3s

iTrader: (3)

you ALWAYS buy the alternator and starter from the dealership, they are the ONLY place that actually seems to rebuild these things properly. actually they are done by mitsubishi, its a mitsubishi part.

Bryan Jacobs

iTrader: (66)

for these I typically stay with either OEM Mazda reman or Bosch

Molotovman

iTrader: (49)

Ask Ray Crowe what he's got in his 20B FD.

texFCturboII

iTrader: (17)

Sounds a bit like this is your problem:

https://www.rx7club.com/showthread.p...t=v8what%21%3F

Have you checked the condition of the power leads from the battery to the starter?

SirCygnus

iTrader: (1)

hm...

is there a ground wire from the transmission to the chasis?

nycgps

iTrader: (4)

First Question : Why are you going to AZ/AAP for starters?

Answer : Cuz it's cheap

and do you not understand this world is all about you get what you paid for? you buy lower/cheap priced products, you get less than desire results. has nothing to do with the country or whoever build it. You can even get crap rebuilds from local rebuilder.

Recently (actaully just 4 weeks ago?) I replaced my Factory Starter on my Rx-8 with a new one, it had about 50K of life (yep) when I replaced it with a Beck and Arnley one. I mean it still works sometimes, but you can tell it's spinning slower and slower every time you crank it, when it was new in 2006 it starts the engines in less than a second, now? 3-4 seconds.

I replaced it with Beck and Arnley because I simply don't trust AZ/AAP's cheap crap, and it works out of box, spins just as fast as the Mazda starter when new. Price? Not cheap, almost double of what AZ/AAP sells their starters.

you bought cheap ****, it ***** you hard, you deserved it.

nycgps

iTrader: (4)

Mazda don't do rebuilds (obviously)

Nor Mitsubishi.

Mitsubishi is just the original manufacture of the Starter (cuz it's all PMG based) and Alternator (Mitsubishi Heavy Industry of Japan, sells all kinds of electrical stuff), they always sell Brand new stuff. not rebuilds

someone else rebuild them, in Japan, 99% of the time will be Nippon Denso.

j9fd3s

iTrader: (3)

they come in a Mitsubishi box and go back to a Mitsubishi address... so yes, its not the car Mitsubishi, but they do rebuilds. i think in the USA.

ND cores get sent to a different place, i will grant you that Mazda's use of ND parts is mostly sensors, which except for the AFM's don't get rebuilt.

BTW the T2 starter N318-18-400R-00 is $168 retail, Mazda USA has 1 right now, you'd better buy it before someone else and makes you wait 6-8 weeks for another one!

Akagis_white_comet

iTrader: (4)

nycgps, just a reminder that flames draw admin attention, especially in a thread seeking information on who does a good job on rebuilds. Not exactly proactive to reiterate the obvious...

I'm well aware of parts store alternators being DOA. Had to install 3 different alternators on my mom's long-gone neon before one worked. It was dead a week later. In retrospect, starter quality seems to be no different because of the country of origin and cost. For all naysayers, they are related because labor is cheaper there, resulting in more parts getting shipped out per hour of labor. Anyway, this is not a thread about the politics of outsourcing labor so let's get back on track now.

Mitsubishi Electric Co. Ltd is who makes the FC/FD/FE alternators and I'd wager the starters as well. They also make things like air conditioners. One pizza place here has had a Mitsubishi air conditioner for over 20 years now. Mitsubishi Heavy Industries makes things like Forklifts and other 'heavy' industrial items.

ObliqueFD & I did a basic crank test with the starter being 3ft from the battery using 4awg welding cables in a direct line. It sounded very healthy, but I couldn't get a rpm reading as the haltech wasn't in the system then. So another battery of tests tomorrow, adding each link back into the chain, starting with the bus bars & haltech. I'll report back with the results as they come in.

korndog517

iTrader: (1)

Well from what this sounds like, is there anything on the front of the engine that could possibly be binding up, i.e. Water pump pulley, P/S pump, etc... I have also seen this happen on other vehicles with an extremely over tightened belt making it to difficult for the start to turn over the engine.

Try spinning the engine over by hand, you should be able to with a bigger ratchet and socket on the front of the "Crank" pulley with no issue. If you feel a BIG resistance start checking accessories

Edit, If its possible to see what the Amperage Draw is on the starter when trying to start would be VERY helpful in diagnosing this, just FYI, if its a stupid amount of amperage draw alot of the times its an actual mechanical issue

SirCygnus

iTrader: (1)

im willing to bet the transmission isnt grounded.