Thank you for teaching us. Because of you, I am launching a business doing advanced automotive diagnostics. I also repair and build PC's and at one time was an electrician. I wanted to expand my horizon and being in electronics and electrical my whole life, it seems like it just fits. I've been slowly acquiring all the diagnostic equipment I see on your channel to get me started. The equipment is very expensive but at the same time no one around here does the kind of stuff you do. I've seen people getting their car towed a hundred miles away to a dealer and fork out thousands of dollars in repairs when I could have went to their site to diagnose and potentially fix their problem. I also want to get into lockouts and all keys lost. I had several people over the last couple of years working at walmart come up to me frantic trying to find their car keys. This kind old lady had lost the keys to her newer Lincoln 3 times and had to have her car towed to a dealer costing her several hundred dollars to get new keys made, when I could have provided that service onsite. Because of you and other youtubers in your position, have given me a new lease on life. It's something I really want to do and being encouraged to start my business is something I needed. Thanks again for what you do showing people like me what its like to do advanced automotive diagnostics and repair.

Use a set of jump leads and make a temporary ground to the engine. Rule out your bad ground.

"What would you do next?" Me: Take it to Pine Hollow Auto Diagnostics 😅

I worked at a Subaru shop (shame on me). The PCMs would fail in unusual ways sometimes. I would run the car with the battery maintainer set to 14.5 volts and keep the alternator unplugged. If it still acts up the pcm might be at fault. I would also check the battery ground to the body. A bad connection at that body bolt will cause all kinds of problems. One more thing to check are the two small braided ground wires from the valve covers to the body that are almost always rotted off.

I worked at a trans shop for 10 years and it was pretty common to see 1998-2006 range dodge caravans that would shift erratically and actually upshift through gears sitting at a traffic like stopped and speedo would rise up to like 30mph or so. It was caused by bad factory alternators putting out excessive AC voltage and the output speed sensor would start picking up this AC voltage and make an erratic signal. We would unplug the alternator connector and it would go away, so we would install a new alternator and good to go, but it was the FACTORY alternators that would do this.

Reliving Saturday afternoon matinee as a kid 60 years ago ! The passenger train is heading to broken bridge ,our hero Ivan is in the cab fighting valiantly against the bad guys and the cinema screen lights up TO BE CONTINUED..........ARRRGH !!!!

My thought is poor ground/s some place. works ok at low current, but when the alternator starts trying to charge the battery the poor ground cannot handle the current. The slow cranking also hints at a poor ground or/or bad battery.

Ive seen that several times on several different makes, whenever I get a car doing something weird I usually try unplugging the alternator first just because it's so quick and easy and AC voltage being put into a DC system can do crazy things to electronics!

I’m suspicious about grounds too.

Make your own alternator harness with test lead, indicator light and disconnected wire to the ECU. If everything returns to normal, that's where you need to look for the problem.

Wild guess - use your battery maintainer to raise the voltage to 14.2 or near that without the alternator connected. See if that duplicates the alternator connected symptom. Guessing at the age of the car there is an electrolytic capacitor breaking down in the engine computer (or elsewhere) when at full running voltage.

Hello, It looks like a ground problem. It should be 2 ground wires that connects heads with the engine block(under the car) and a ground wire that connect engine block with the battery and body of the car. Check that wires.

Case of alternator not a good enough ground to engine block

I had a bad battery cause symptoms like yours. It made the alternator go to full output and would overload the poor little 4 cylinder. It was a huge tractor battery that the owner installed.

my next experiment would be to remove the belt that runs the alternator and run the engine with the alternator unplugged and plugged in and see what happens. Another would be as other said keep the alternator unplugged from the vehicle and manually turn the alternator on with some kind of set up and the opposite of this. Have system voltage maintained by the maintainer and crate some kind of set up to plug into the harness and fool the vehicle into thinking an alternator is attached.

i swear i like this channel more every week.he gets into some weird rare problems, but finds solutions to 98% of them.gotta respect such dilegence & determination.

I have had two alternators that had corrosion on the ground stud that mounted the alternator's voltage regulator. I pressed out the stud, cleaned out the corrosion, coated the parts and stud hole with aluminum wire contact paste. The problem of lights dimming and brightening went away. Maddening, as the problem was intermittent, road bumps, rain, hitting the brakes and just turning the wheel at times. But when I would check it, it produced the correct voltages.

200K+ subscribers and I don't think you've asked for anyone to subscribe or even *"SMASH THE LIKE BUTTON"* !! You're definitely doing something right!!

Need to confirm and diagnose the parasitic first. If not related to the running rough issue I'd pull the accessory belt first. Then test belt on with disconnected alternator output wire. My guess is some sort of bad ground. Take a meter and measure from the alternator case to the negative battery post while running rough. That will tell you immediately if there is a floating ground.

Ivan knows how to make us anticipate his next videos, & it hurts 😢

Try a jumper cable from alt housing to battery ground and see if the problem goes away. I suspect a missing ground.

I am going to bet that the parasitic draw has something to do with it since it is a B+ circuit. Shorted wires.

That alternator wiring diagram is what I've wanted to check on a Toyota Hilux 1990 alternator. It produces 13.4V max, so bought all OEM rectifier, brush kit and the IC regulator. Before I install the IC regulator, I wanted to check the existing IC regulator by applying batter voltage to the battery wire and check the voltage at the brush kit (while it is not installed). I learned a lot from your channel and wanted to say THANK YOU IVAN!

Perfect release timing before date night! Thanks Ivan! 🇺🇸

golly.... what a great channel... and a great mystery.... Just started watching and had to jump in...

I had the same type of fault ... same car ..... new alternator fitted .... brought on engine management for air flow meter...... turns out only original manufacture alternator was the way to go ........crazy never seen it before lol

I've scoped those coils before there is power and ground and the third one is a pulsed close to 5v square wave that signals to the coil from the ecm to fire.

Sounds to me like the interior and engine bay are low on gasoline.

One of those wires on the alternator connector is shorted to a coil or injector wire. Bypass the connector and manually run a length of wire directly to the battery's positive and another one to the battery's ground. Plus, that connector looks like it has three wires!

I love your videos because I know if anyone can figure it out, it will be you! Waiting anxiously for part 2 Ivan!!

banger of a case study this one. cheers for sharing mate!

Sounds like Subaru hired engineers from the HIT & MISS ERA.

Very intriguing / challenge ! Looking forward to seeing / learning more !! 👊🏼 🔥

I had a Chevy Chevette. years back. Ran fine started, charged fine. However once in a while on a damp or rainy day it would not crank. In the end it turned out to be a poor starter ground. Aluminum starter housing on an aluminum bell housing. Sanded clean, issue solved. Grounds can be confusing. Some cars , as you know ground the battery to the chassis instead of the engine, then run a bonding jumper to the engine. The flexible braided ones look OK, but, are great for turning to dust her in salt land.

Wow! Put crazy on it, Ivan! My first though was that the alternator was loading the engine excessively and whatever other problem the engine got, was increased. But the possibility of a bad ground being overloaded by the alternator seemed even more plausible. Maybe it's a mix of both. Can't wait for Part 2!

Thanks for the video Ivan.

I'm racking my brain, but I can't even put together enough of a hodge-podge to speculate! Dang Ivan, making me wait for part 2! This is a spicy one~ :D

Man thats cool! Im curious to see what you find.

I would do two things. 1 determine which coils are dropping out and 2 find the parasitic draw. Look for common interface. What is pulling the power away from the coils. It would also help to know if this problem existed with the original stock alternator.

Back in the "good ole days" when you could actually rebuild an alternator where voltage regulators were external of the alternator; crazy things happened when the bridge or voltage regulator failed. In those "good ole days" if an alternator failed, the battery and starter usually wasnt too far off from failing also - but you couldnt sell a battery and starter as preventive maintenance to a customer. When battery or starter did fail it was your fault because you fix'd the alt and or parasitic drain. Hence, in the old days you would write your recomendations on the R.O. what the customer might experience. You said earlier this was an aftermarket alternator, we've all bought aftermarket alternators w/out issues - but not knowing where it was bought is an unkown variable. Before Autozone or O'Reilleys there was Chief Auto Parts. Chief had the alternators and starters rebuilt in Mexico; the prob with their rebuilds is they only replaced the broken part and never did a complete rebuild. It was normal to go thru three alternators or starters before you finally got a good one. So when Autozone came along Chief Auto was pretty much put outa business. Autozone, and O'Riellys, completely rebuilds their units atleast they did early on. Now we have to deal with Dorman, Cheap Korean stuff, and cheap Mexican stuff. Dont go cheap when buying Alternators or Starters: and get the lifetime warranty. On my vehichles I always always fit a 1000 cca battery when the old one fails. Sorry for the long post. Im going with a faulty aftermarket alternator or a broken crusty ground in the Alternator circuit. BTW, I've had alternators that would not act up on me over time left someone stranded. I still remember a Lincoln that did this three times. Always acted fine after removing it to be bench tested - it always tested good, heat related (who knows). Finally talked the owner into "just replacing" the alternator as every piece of data kept pointing to the alternator. New alternator fix'd the problem.., was this a win for the Parts Cannon😂

Cool! Still suspect a short to ground somewhere on the main charging wire. Something is loading that alternator down pre fuse. Love this stuff! 👍👍🇺🇸

I would check the grounds you were mentioning. Subarus are notorious for weak grounds when they are all good. I guarantee the ground straps on the bottom of the engine are rotted off. So the ground to the starter is the only ground you have. I bought a 08 Impreza for $500 because they couldn’t get it to ever charge the battery. New battery n starter. Still no charge. I got it home and checked the grounds. Straps on bottom broke. Ground at starter loose. Fixed the grounds and everything works fine now. Oddly though it would start just fine so I guess the starter doesn’t need much ground to work. They do spin over pretty easy.

Thanks Ivan. No idea what's next, but as a former (happy) Subaru owner, I'll be surprised if it's not something fairly simply causing the erratic behavior. Can't wait for the next video.

Thanks Ivan!

I don’t know why resistor pack comes into my mind or if this year of Subaru has one but I’ll guess it because you’re Ivan and I already know it’s fixed. Great video.

I would check the charging current when the alternator is plugged in, it sounds like it is trying to charge very heavily. Love your channel, wouldn't miss an episode, thanks.

Next thing I would try is disconnecting the main output wire from the alternator, connecting the battery maintainer (set to around 14 V to simulate alternator output) to the disconnected wire, and see how the engine runs with and without the alternator field connector plugged in. How much current is being pumped out from the alternator? I would also take a minute to look into the parasitic draw complaint--it may be related?

Also, this just popped into my mind. If you can check the voltage on the fuel pump with the alternator plugged in and not plugged in and see where that takes you.

Ivan, there is always a look of delight when you say something is crazy. It's kind of like Dr House saying "interesting!"

Replacing the alternator cured a rough idle on my Volvo V70 D5. It also transpired that one of the ABS sensors was unplugged, but only threw a code after the new alternator was fitted. Most strange 🤔

Crazy case Study I'm still suspecting a bad ground somewhere!

I had a similar issue on a subbie, the cause was a poor ground at the ECU. The MOSFETs that drive the injectors are working at 5V. When the battery voltage went above 12V and because of the high resistence of the ground on that voltage regulation circuit, it would create some sort of a weird RC feedback and thus mess up intermitently injector drive. Next steps would be to check powers and grounds at the ECU and check for good injector drive with a noid light.

I've never seen anything like this either. Yup, mind blown. That's nuts! But something tells me we are having a poor ground issue here somewhere.

Subaru thinks its an Isekai protagonist. It's in another world.

I installed a Subaru 2000 2.5 in my 1986 vw vanagon I’ve been fighting this high idle stays at 1500 rpm’s no matter what I do. Replace it did drop when I replace throttle body and rpm’s did drop but didn’t last i do have the Subaru harness on this swap thanks for the awesome videos you.

I had the same problem on a Ford many years ago. Problem was a bad diode in the alternator causing high ripple on the 12V. ECU did not like the ripple and went nuts. New alternator fixed the problem.

I'd check the grounds, including ground cables. Once had a battery main ground cable that was corroded in the crimp fitting. Took forever to find it.

Ha ha, you left us hanging. Along the line of gaining info, I think I'd look to identify which coils are dropping out. If we can show it's the same coils each time, we could swap position of the coils and see if the abnormality follows the coils or stays with the driver. And while at it, I'd try to look at the ground wire for the specific coils that are abnormal. Perhaps they share a branch ground and the connection from there to the mother ground is tenuous, exposed with more power use. What a great case....

Everybody's different, so are everybody's troubleshooting techniques. upon discovering that connecting the alternator cable resulted in rough engine performance, I would have de-pinned the alternator connector and then reconnected the connector pins individually one at a time to understand what wire, when connected, caused the rough engine performance and then followed that diagnostic result.

What I would do next is to put the Tornado 5000 at 14,7v with the alternator disconnected and see if the problem is related to battery voltage. If that does not cause the same issues as the alternator, I'd first put the ground terminal of the Tornado to the alternator case, and if that does not reproduce it put the positive terminal of the tornado to the positive terminal of the alternator. Basically let the Tornado simulate the alternator. If even that does not reproduce the problem, it's time to remove the belt from the alternator and see if the problem is still there if the alternator is only electrically connected.

I'm going to throw out a pretty vague guess as to what I think it might be. There is a wire somewhere with failing insulation. With the alternator disconnected the power flowing through it is low enough to stay isolated and not cause any problems. But as soon as the alternator gets plugged in that faulty insulation can't shield the wire anymore and it's getting grounded out.

Interesting for sure. Back in the mid 90's I had a 1990 Corvette that did something very similar, and it was an injector with resistance way out of range. Don't remember the details, but it would barely run with 14V, but would run ok with 12V or mid 11's.

The only thing I would do next is, check the oil.

I haven't finished yet but you could mimic the alternator with your Top Don power supply. Connect positive to the alt output and ground to the alt chassis or closest equivalent ground point. Turn power supply on and off and see if the results match. If so your ground theory is sounding very good. Make sure to probe the worst coil and check positive and negative. I would assume you'll find something and then troubleshoot why the positive is being pulled down OR ground is pulled up.

Great analysis. I was ready to condemn the engine ground, especially after seeing the appearance of the battery connections with rust on the battery clamp nuts. (Why can't people just clean them and coat with vaseline? My batteries last 8 to 10 years in the Northeast with simple maintenance like that.) But I would have been dead wrong. That's the difference between my dumb-assed guesses without careful oscilloscope work, and what you do.

I would try enabling the alternator by bypassing the connector. Assuming the regulator is internal. You could use your bench supply to vary the excitation and delivered current and note the effect.

Wow, this is your kind of challenge Ivan!

Tanks for sharing, i guess the polarity of the alternator gets backwards when you plug it in do to some short to -/+

What if the coil you're reading the ground on is one of the ones that's still performing? Next, I'd run the same test on the alternator case. This car reminds me of the one that Bernie had that had the ground modified and flowing across the crank. If all other ground checks are good, I'd install a temporary ground from the alternator case to the battery. If that fixes it, I'd be back to looking at the ground path. With multiple ground paths, it could electrically test good, but be flowing on the wrong path.

I'm thinking the alternator AC voltage is somehow contaminating the ignition circuit 🤷‍♂️

Wow ! The previous Work and money spent has to be substantial. Now the back check of all work done just to make sure all the parts hung on it are good and installation Correctly completed. Then .. Ivans own Diag.. This is going to be a good one❤

I bet it has something to do with the parasitic draw on that 15 amp fuse. What circuit is that fuse in, what is the draw, what is the draw with the alternator plugged in and what is the draw with the alternator unplugged. Something is probably jumping to ground in that system

Check out the parasitic battery draw, maybe related.

I would connect a ground directly to the alternator just to make sure it wasn’t a poor connection to the block. Then I feel like at this point I would just power the alternator through a direct power source and bypass the plug. If it charged and ran normal I would also plug in the lamp wire to make sure that wasn’t causing an issue. If that’s the case then maybe a problem with the supply wire. If it was still acting up with the isolated supply it would have to be the alternator.

Alternator is the issue, had this a couple of times. Usually pops a cam code, if you scope the amps output from alternator you will find large negative spike, I guess must confuse the ECU. Good luck. Stu from NZ

I'm waiting for part 3 when the owner sells it and the new owner brings it to Ivan.

I would unload the alternator and verify the running affect continues with the rotor energized or not.

So my first thought at this stage would be to pull the ecu fuse and feed in external 12v DC power ideally from a variable power supply to see if either 1) The ecu is being sensitive to ripple from the alternator, or 2) By turning up the voltage on the variable psu to see if the ecu starts dropping out when it sees 14/15v

That noise and spike level is crazy when the alternator is connected - on both 12V side and coil ground. I'd not be at all surprised if the spikes are causing problems or messing up sensor signals.

A few days ago one of my guys changed the alternator on a Volvo of some sorts Im not sure if we diagnosed it or not, but it was bad apparently. I just walked by as he was starting it up and i noticed there was a new battery on it. Which i know we did not put in. My gut told me to check the amp output on the alternator so i fetched my handy smaller ampclamp which max out at 100 amps. That clamp did indeed max out on the alt. wire. Told the mechanic to put the battery on a charger for a few hours so we didnt burn the new alternator out on the get go. Cause it looked like thats where the majority of the amps went. To the battery that is. So he did. I went over there now and then to check on it and monitor the amps and it turned out the battery could take whatever the charger could throw at it if i turned it up. 60 amps in this case, and it did not go down much in the time we charged it either. The battery didnt even get hot. It was so weird. We put a new different battery in it and it was all dandy. What is the relevanse to this case. Well... lots of amps need a clear path all the way arround the circuit. So a badish ground could have done some weird stuff. I know it started ok but still... That case just popped up in my mind when i saw this vid...

FYI... Interstate batteries aren't what they used to be. Had 2 bad ones in a row. They were my go to until 5 years ago when quality went to sh*t.

Cranking sounded weak to me so I’d check for poor engine to chassis ground. With the alternator working, the engine ground could be getting raised. You could use the charger with the negative connected to the engine block and see if it does the same thing.

Raise the system voltage while running using the maintainer to alternator levels. Check cam sensor signals. Jump the battery ground to the engine. Do a voltage drop test to the block during alternator connection. Not in that order.

Mind blown as soon as the alternator was reconnected.

Bypass alternator plug (go direct to alternator) white/red with fused battery voltage and test light for alternator warning signal to isolate problem to vehicle or to alternator output.

Alternator charge wire hooked to battery negative creating electronic breaking. or plug pushes the pins down into the commuter jamming it. LOL That is wild......

More like Crazy Ivan fair play for taking it on what an interesting issue

Bad diode or short in the charge warning light circuit? Lots of scrolling on my phone, but based on the clues you revealed - 1 reasonable DC output 2 parasitic draw 3 runs well when 13.6V @ 18A applied directly to battery. So i looked for diodes in the circuit between the Alt and a 15A fuse - "combination meter" charge warning light. If correct, who'd think to link the cluster with an engine running on 2 cylinders? Great video as always Ivan & thanks for the puzzle!

As an electrical engineer, I would say before you pull out all you equipment get your Fluke out and measure the ground connections. Shouldn't take > 5'. After that you can make video #2. OK, we only get to see what Ivan wants to show us, but still, coming from the rust belt I'd say definitely ground check. And why are the copper windings of the alternator black? New color scheme? When I played with that stuff, only burned up transformers would have black wiring.

Shorted diode in the alternator probably.

Interesting case study. Is alternator on a CAN-bus? Is it possible alternator is being commanded high amps when engine at idle??? Could remove belt and confirm. Had very different issue, but related to in-vehicle network with my BMW E60. Alternator not working because CAN-bus pulled down by faulty oil level sensor.....grrrrr! Waiting for final outcome. Great video as always!

I did a search and found a strange solution. A subject with a GMC had the same issue and it turned out to be mismatched injectors. When the alternator was plugged in two injectors started flooding the cylinders causing those cylinders to misfire; he theorized the additional voltage caused the wrong injectors to inject too much fuel. Replaced the injectors with the correct ones and the problem went away.

What would I do next ?? I’d wait for part two and learn from Ivan as he always figures it out without me 😉

Maybe coil power shorted to field power? Maybe see if they are supplied on sep circuits, and if so measure current on coil circuit.

Greetings from Australia I would be tempted to put a capacitor on the output to GRD of the alternator just to see if the AC ripple is causing the issue .

Do a wiggle test the check coils.

always start at earth connections seen wierd things with bad earths also trailer plug bad wiring and or aftermarket things like led lights,stereos,speakers etc etc

There's got to be a bad earth possibly on the ECU as there doesn't appear to be a problem with earthing to the Coils, Something has got to be floating as soon as there is any ac ripple or once there is a significant current flowing into battery.... however, I would also look into the current drain to identify where this actually is - I wonder if they could be linked.... (interesting that the engine gets progressively worse when alternator initially plugged in. As well as the 1st good current ramp takes 3 more steadilly increasing ramps before getting another good pulse) First check the basics - Power and Grounds plus identify where that parasitic draw is! If the signal are getting weaker when Alternator is plugged in, thats got to mean the current draw into battery from Alternator is causing a Potential difference (something is floating) across a bad joint somewhere (Power or Ground but most likely Ground)

my bh5 did that , My cam sensor signal was getting messed up only on start up , thought the alternator caused it, as did not fault with alternator disconnected, checked the grounds and cleanned up and runs good now

Coil ramps when alternator plugged in suggest a fluctuation in amperage which suggests an ac fed source versus dc and returns to nominal when the ac curve cycles. Either that or ground isn’t allowing higher amperage when plugged in causing the system to go wonky.