Did I MIS-DIAGNOSE This Mercedes?? (GL450 no A/C)

[@suryaprihadi2258]

Ivan, when you do HVAC , first thing you need to verify is the STATIC refrigerant pressure when engine is COLD, where all components are at ambient temperature. Minute 1:01 shows a very low R134A pressure, which is 2.5BARg or 36.3 PSIg. You then take a R134A temperature pressure chart to see what is the actual pressure/PSI to be at "X" Fahrenheit. Now, this GL450 has a whacky ambient temperature sensor because it read 125.6F at minute 1:19 while interior temp sensor at 1:01 was showing 79.7F. As per R134A press/temp chart, if at ambient temp of 80F, the R134A pressure STATIC and compressor was not operated the last 12+ hours where between LP and HP circuit it would have equalized and all back to ambient temperature, and that should be at 86.7 PSIg. So if the refrigerant pressure sensor is not defective or drifted, this car has too low a refrigerant pressure and the HVAC computer will not allow compressor to run and may not produce any DTC too. 36.3 PSIg correspond to R134A when at ambient temp of 41F or 5C, which can happen in a cold country and hence DTC may not be triggered because HVAC computer does not think it is a faulty value. Now the MB protocol itself. These are components which will and can prevent HVAC compressor from running, if their value is whacky : 01. Ambient temperature sensor, usually located at the front bumper. 02. Interior temperature sensor 03. Refrigerant pressure sensor 04 Engine coolant temperature sensor The radiator single 800 Watts cooling fan called M4/7 is a shared cooling fan between the engine and the HVAC. You can see under the Engine Computer module how many % is for engine and how many percent is for HVAC. 10% duty cycle is OFF and 90% duty cycle is maximum for this M4/7 fan. The initial and last 10% is for diagnostic purpose by the engine computer. Loose/disconnect the PWM sensing wire to M4/7 fan, the fan will kick in at 100% full speed as precaution, so becareful where you finger is at when you access the fan wiring. You can not remove the M4/7 fan fuse, it is not those plug-able fuse. It is at the F32 Front Pre-Fuse box and those dumb-azz special fuses they use where they combined like 4 fuses to 1 strip. According to the VIN number of this GL450, its HVAC system is called Code 580 by Mercedes, meaning a 2 zone unit and not the higher-end version called Code 581 or dual evaporator called Code 582. What bothers me is that your thinktool showed at minute 1:01 rear evaporator temperature of 189.5F, this car by VIN number does not have a rear evaporator (2nd evap ) so there should be no temperature value or scanner must declare NOT EQUIPPED. If you do not have MB WIS/EPC, this link has good enough VIN DECODER and is accurate enough for this truck : carinfo.kiev.ua/cars/vin/mercedes/vin_check?su=vl00tk8r0dpy Always get the Data Card of MB cars when you are working on it for equipment like HVAC, as the options are plenty and you may ened up using the wrong wiring schematic between Code 580 and Code 581 and Code 582 ( with rear evaporator ). The MB original wiring to use for this car is HVAC computer is : PE83.40-P-2101-97maa Wiring Diagram for automatic air conditioning (AAC) control unit Model 164 except code 581. Comfort automatic air-cond control unit N22 N22 is the HVAC computer module ID as per Mercedes WIS or Xentry scanner for this car/truck. N22/7 module is mainly for W212 E-Class. If you can, avoid using re-drawn diagram from even the best of ALL DATA or whatnot and more for if for Front SAM N10/1 10 pages, which even the MB original one is lacking of update for my W212 2014 facelift and E400 M276.820 is a pariah engine variant very uncommon. Typically MB uses Denso compressor and there are times where if the compressor replacement is non-genuine and has a much different electrical operational value of the control valve, DTC may be triggered. Example like what you get now defined as open or short circuit 9006 / 9007. Here is a good example of 9006 DTC solved by going genuine Denso mhhauto.com/Thread-AC-NEW-COMPRESSOR-WILL-NOT-START-ERROR-CODE-9006-HELP-ADVICE-PLEASE?page=2 Now the GROUND wire which confuses you. Often MB separate & later combined their grounds for certain components/sensors. For this compressor control valve, it uses the Z50/37 HVAC Comp ground connector sleeve and it is also connected to pin #1 ( describe as a 31) connector D of the N22 HVAC computer, where the final ground destination of this wire is ground stud called W29/2. W29/2 is at A pillar, the right side one. "W" is always a ground on MB cars. "Z" is almost always some sort of splice. 31 is German DIN speak for ground or return path. 30 is always HOT , +12V power for MB/German cars. 30G is +12V but via a big relay af the F32 Prefuse box, so it is switchable but by ECM only. Sometimes the pin at modules has real Signal Return description, usually for the 5V sensors return signal. It is common for MB to make sure the small component/sensor ground is not placed at anyhow anywhere there is a metal part of the body, NOPE they do not do that. They would want their small component/sensor ground to share the same ground wire and stud as the module, this is for feedback/diagnostic function because that is the cleanest path back to the module. This GL450 would easily have 50+ modules, my 2014 W212 E400 tropical country has 47 total modules, and 4 seasons country and an SUV body usually have more modules. This very plenty of modules is why MB wants their signal ground to be closest to their power ground for that module and components/sensors supported by that module. You do Pico scoping often, you know how noisy is a ground is ....electrically. I hope this information will help you when you visit the GL450 next trip.

[@tomtke7351]

wow

[@janvanderwal2744]

Wow 2

[@mdushaka6403]

This is a genius 🤞🤗

[@mdushaka6403]

This guy is a genius, all this information for free🤔🤔

[@PineHollowAutoDiagnostics]

Amazing info :)

[@dustcommander100]

The valve or coil in the compressor differs from the test loads in that it is inductive. You will probably see spikes on the feed from the controller when the valve is connected - but that should be the same when the body ground or compressor ground is connected. I'm wondering if the ground at the compressor is good or if it is being pulled up because of other loads from the engine having higher currents needing to go to ground. Can't wait for part 2. I know you will get this!

[@g1jee]

The valve coil is an inductive load, bulbs are purely resistive. I would guess the flyback diode in the compressor valve circuit is the culprit. The job of this diode is to eat any kickback voltages when the valve is operated or released.

[@major__kong]

Why do all the videos I'm watching this morning somehow involve flyback diodes? Over on Rainman Ray's channel, he's pulling the alternator lead off the battery while charging and somehow didn't fry the thing.

[@Stoney3K]

It's a brand new compressor, so I wouldn't suspect the flyback diode to be bad, unless the flyback diode is on the HVAC controller -- which is terrible design. Some filter capacitors on the HVAC controller board may also be causing the problem.

[@EATSLEEPFARM]

@@Stoney3Ki think there is two separate compressors that look identical but one has the diode others do not

[@Stoney3K]

@@EATSLEEPFARMIf that is the case it should be easy to measure. From +ve to ground it should read the coil resistance (or open when it's on diode measurement), and from ground to +ve it should read 0,7v in diode mode. And in that case the fix is just as simple, adding a 1N4001 in the reverse direction would be enough.

[@EATSLEEPFARM]

@@Stoney3K watch part two its out 👍

[@sprograt]

Good Morning Ivan just flicking through YT and a new PHAD video popped up, Thanks again.

[@Vespastendert]

Best feeling ever

[@simonparkinson1053]

Hmmm Same codes with the original compressor? I'm wondering if the original compressor had a back-EMF diode built in and the aftermarket one hasn't. Therefore the driver is happy with a non inductive lamp load but doesnt like the spikes when it's PWMing the compressor coil

[@randy1ization]

no telling where the guy got the replacement compressor.. that might end up being the rabbit hole..

[@GarnConstructionInc]

Cool Video! Time to get the Pico on the case! My guess is the compressor valve signal ground noise. The computer receives inrush spikes when the ground varies causing a fault. A good test would be to add the body ground to the connected signal ground.

[@z00blez321]

Dude I absolutely love your videos. Been watching for years. You've made me such a better diagnostic tech. Can't thank you enough.

[@kevin9c1]

You need to follow that ground circuit. It may not go directly back to the HVAC module. It may go to the front SAM. There could be more in the loop than just the HVAC module and an engine ground. Other note is that I have noticed that when you turn the A/C off on Mercedes, it doesn't immediately shut off the compressor. It takes a few minutes.

[@sasquatchrosefarts]

First thing is to clean and test all harnesses with a test light and probe. That's diagnostics 101. And clean any ground bolts to the frame. Yes, it wastes a bit of labor if that wasn't the problem, but corrosion is funny it behaves differently depending on air temp and humidity. I had a bad computer ground bolt that was tossing spark plug wires off the distributor on a 1994 Volvo. Electrical shorts and overloads and interference causes all kinds of funny stuff. KZbin is severely lacking on systematic electrical system cleaning. It's worth it for this vehicle owner to spend $300 for someone to literally clean all harnesses and ground bolts in the whole vehicle. Then you can approach electrical systems much more systematically.

[@sasquatchrosefarts]

And it might only take an hour for a good.mechanic. one hour of labor to restore all harnesses and bolts to clean and low ohm status. And it must be done every ten to fifteen years anyhow, on any vehicle .every single connection. It's like changing the oil. It has to happen every decade.

[@PineHollowAutoDiagnostics]

@@sasquatchrosefarts I disagree...First thing is to DIAGNOSE the problem before touching ANYTHING lol!

[@willypalooza6519]

​@PineHollowAutoDiagnostics ive alway been taught and practiced dont touch anything before diagnosing it. Dont even touch the gas cap even if the code is a gross leak. Find the issue. And confirm. If some kind of issue is still present. Dive back in

[@timothyhults2549]

I would assume that the ground wire must go through maybe the ambient air censor or a bad ground at pcm or ecm tough diag. But persistence would be the key.

[@Stoney3K]

I would first suspect a bad controller ground. If the fan speed is ramping up randomly and it signals a shorted compressor if it's not physically shorted, it would suggest the voltage sensing is messed up somewhere. The short circuit detection is probably done by measuring the voltage drop across an internal shunt resistor, if the ground is bad, the controller would detect the shunt as being either at +12v (open circuit) or a virtual 0v which reads as a short. Compressor ground looks fine if it can carry an amp of current, but the ground path going from the compressor ground to the HVAC controller ground may be bad.

[@nevillegoddard4966]

@@Stoney3K Yep, that seems to make sense!

[@fredmalito9869]

I’m curious about a bad engine/to chassis ground, compressor to engine ground, or maybe something leaking positive voltage to the engine ground path, such as the alternator.

[@namvet-l8m]

3:22 Should go to the negative battery terminal. Should check compressor/body/engine to negative battery terminal continuity. I bought a battery checker and attached the + to the battery and - to bolt on top of strut and voltage was 11.8 volts. Connected - to - battery terminal and read 12.9 volts! Fixed that ASAP.

[@abdouallahuniverse5689]

The AC compressor is easy to test with the car off before replacing it, just remove the connector and run a direct electric wire from battery to the compressor, positive to compressor connector and ground to compressor body, and see if the clutch engage or not

[@PineHollowAutoDiagnostics]

Ummmn did you not watch the video? 😅

[@AutomotiveEvangelist]

My brother has the same car, same year...now driven by his daughter. It's having similar issues. I diagnosed blend door actuators over the phone (different temps coming out of different vents.) I used to own one of these for years and they are great road trip cars, so hopefully your customer enjoys it once you sort it out.

[@RussellBooth1977]

Even though the air conditioning compressor clutch can be energized manually I would still check the compressor clutch coils Ohms reading & compare it against specifications so as not to take anything for granted meaning that the compressor clutch coil could be overloading the module due to a short circuit inside of the coil or the Ohms reading could just be too low. I think that I mentioned that a faulty ABS module was causing interference with the body control module which was causing my 2002 model Holden VY SS Commodore not to have air conditioning sometimes. As far as the 0.5 Meg Ohms reading apparently it does matter if the multimeter reads that from the secondary ignition coil post to ground because I rectified that problem in my fathers 2000 model Holden VX Commodore yesterday. It runs the Holden ECOTEC version of the 3.8 litre Buick L36 V6 engine & there was Infinity Ohms reading between the secondary posts of the middle ignition coil for cylinders 2-5 & about 35 Meg Ohms reading between the secondary post of cylinder number 2 & ground. My point is that could appear to be a short circuit at about 20kV so it was tracking to ground,all the other coils measured Infinity Ohms to ground, cylinder number 5 also had a strong blue spark but cylinder number 2 had a very weak spark. There happened to be an old Buick L27 engine sitting in my dads shed which came out of an earlier 1993 model Holden VP Commodore which my youngest brother crashed into a power pole so I pinched the 2-5 ignition coil (same coils) off that engine & fitted that into my dads VX Commodore which fixed the problem. Note that the other 2 coils read at 6k Ohms between the secondary posts which is good !

[@moshet842]

It has no clutch I believe.

[@leeross7896]

google search @11:52 "It is important to note that Mercedes has a 'special' Control Valve that has an internal diode - many aftermarket control valves and aftermarket compressors might come without the diode, causing a trouble code DTC 9006 "Component A9 ( AC Compressor ) has Short Circuit" error, and the HVAC controller will shut off the"

[@juancarlosmorales6775]

Depends on the platform. I have 3 MB without the diodes on the AC compressor, and 1 with it. Go figure. It took me some reading to find out. Denso provides both types of setup for applications

[@kerrylewis2581]

It looks like the GL is in nice shape. I'm surprised the owner waited so long to get the A/C fixed.

[@destruct61]

He was probably like me afraid to hear the problem and the price lol

[@roadshowerun1770]

Just an observation about light bulbs... Incandescent bulb are not a direct substitute for a solenoid when it comes to current measurement and testing. Incandescent bulbs have less resistance when cold. Tungsten filament cold resistance is on the order of 10x less than it is when it heats up. They heat up fast, 10-30 milliseconds. But they can present a challenge to power management. Ask anyone who managed theater lighting... lol Solenoids have an initial resistance to a change in current. Once the field is created (and current is flowing) they have a tendency to spike with a back EMF when turned off. The peak of the spike (the voltage rise) is dependent on how fast the current is turned off. Clipper diodes (reverse bias configuration) snap on fast (10-50 nanoseconds), and if sized correctly can catch most of the back EMF. If their duty/heat/current rating is a little "light" they can self-destruct. Ivan's circuit confidence diag of the controller wire appears to demonstrate the controller can output 900 milliamps without any problem. It would have been interesting to run the light bulb string from battery positive to the compressor controller pin and measure the voltage drop across the light bulb(s). Would the compressor come on? Are the measured voltage drops across the bulbs and the compress valve making sense. Yea, if it were me I would using a 20 watt/15 ohm power resister instead of a bulb. I have more to them sitting around anyway. I don't think the "Scotty fix" would work. Not with a single overlay. If the compress can be driven directly from battery positive with an overlay, you would still need an overly for the AC controller... since it is measuring/monitoring current... The 12.7 ohms from the compressor to chassis ground should rule out an open or short on this leg of the circuit. When driving the controller with the light bulbs to chassis ground it appears to be happy also. I guess you should check the controller ground points? Interesting the controller is not using ground side (shorting) to manage the compressor valve. How does the compressor valve find ground? Is it possible someone reversed/flipped the connector at the compressor and the controller is being driven into a dead chassis short/ground? Love the content Ivan. I never miss an episode

[@jws3925]

You do more one handed than a lot of mechanics with two

[@mph5896]

He has alot more time to do it as well though. He may spend a day or two on a complex diagnosis job. Most shop mechanics get about an hour and have to move on.

[@turboimport95]

@@mph5896 this is 100% true unfortunately. Flat rate guys don't get the same luxury of time as he gets, As he works for himself and makes 100% of the money not 20$ etc for 1 hour. That's why a lot of shop guys throw parts. They need to redo the whole flat rate deal because it hurts the customer more then anything.

[@rj.parker]

He does more by himself than some shops with five guys. He bails out many of them especially on electrical and timing issues. An engineer who knows physics and wrenches.

[@turboimport95]

@@rj.parker no one says he does nothing.. We know he does a ton for the shops. But it still boils down to the whole flate rate issues that exists right now. If i work on my own cars i can spend 2 days diagnosing a issue. Guys in shops cant do that.. They wont get paid..

[@PineHollowAutoDiagnostics]

@@mph5896 Yes, complex diagnostic jobs can eat up a lot of time, even while working efficiently. Good thing I charge by the hour LOL!

[@topher8634]

The lamp is a resistive load. There's no inrush when energizing it. Magnetic clutches have a diode across the coil to absorb the back EMF. The solenoid valve is an inductive load like the mag clutch. Maybe this one does too and it is bad. Look for a diode in the fuse box or in the circuit.

[@wallebo]

You, sir, are next level for sure. It surely wouldn't take long to call roll.

[@mikere5251]

I change just solenoid from Amazon (62$) and no problem. I had exactly the same problem on the same car.

[@aaronsmith4746]

The inference is this is a compressor without a clutch, and the control is to a solenoid that changes the compressor output. I just learned about those systems with my old Lexus. The control is pulse-width modulated, so the scope would have been interesting. There are at least three connectors in the electrical path to that solenoid that could be checked. Adding the resistive load of the light bulbs in series with the solenoid might trigger some threshold check for low current in the controller, which it reports incorrectly. The controller may be evaluating the switched current flow through the solenoid as it does PW modulation. It should be able to tell if the solenoid moved in response to the current. Maybe the solenoid is mechanically worn and isn't moving quickly enough? Although it seems like that could be a problem with a pure resistive load too, and it was happy with the light bulbs.

[@jeffdickeson1497]

I know this is an old post, but this may help someone else in the future. The compressor used here is the 7SEU17C which is used on many different vehicles of all makes and models. The solenoid that is on the compressor that Mercedes uses on most of their applications has the diode built in while most other solenoids by other vehicles doesn’t use the diode. So bottom line if you order this compressor by part number you have to make sure it has the correct solenoid installed on the compressor for a Mercedes. If not the computer will disable the a/c.

[@sebry25ify]

We have a ground issue . Your measuremnt were based on body ground not engine ground. I am saying based on what I learned on best youtube channel when it comes to circuit testing staring Ivan.

[@RJon2006]

Part 2 should be good. Everything is complicated and junk. Love this channel!

[@norcal715]

Cmon, Don't leave us hanging Ivan!

[@vwwrenchie314]

COOL.. Very COOL nice fix Ivan..you are the professor of that compressor 😎👍🇨🇦

[@hughkoontz2166]

Wow Radio Shack😊 Havent seen anything from there in decades. Amazing fix!

[@theadmiral5425]

The a/c still works....its called 4 x 55. A lot cheaper than throwing both parts cannons(new & used) at it.

[@jessicav2031]

Sounds a lot more like a bad ground. The test lights work with one ground but not another? Not sure what is ambiguous about that. If the engine has a high resistance to ground it will be sitting at a much higher voltage than 0v, so when the HVAC computer looks at the undriven circuit it will see that and perhaps think it is open.

[@moshet842]

Well, the issue in this case is not that the test light works with one ground but not the other. It is the HVAC controller shutting off the valve activation signal when it detects a current draw that it deems too high. The test lights are in line with the valve activation/driver wire. You might be right about ground, however, why does it work when he uses only one bulb?

[@Anonymous-zv9hk]

Heat Index on the coast of SC has been up to 110F, if not higher, over the past ten days or so.

[@hightttech]

Good cliffhanger 👍. Lamps resistive, coil inductive; maybe a clamping diode in compressor funky? [breaking down prematurely, or missing?]

[@htownblue11]

Oh boy. This one is going to involve a deeper dive into the ground circuit on the inductive coil valve.

[@JoseSilveira-newhandleforYT]

This one not so easy, Ivan! I would check continuity of the grounds under load and check for spikes in the compressor - any of those may send an overcomplicated control system to uncharted territory. Very curious to see what happens next 🙂

[@john7017]

Try testing the compressor regulating valve with a Megger tester to check the integrity of the coil windings. If the windings are breaking down, not a dead short, you may get a bleed thru that the controller doesn't like.

[@juancarlosmorales6775]

Ivan, if you have the car again, would you mind showing the top of the electronic control valve? If it has a yellow dot mark, or a triangle stamped on, it has a DIODE built-in. This diode is required in some MB models, but not all of them; therefore, did the old compressor has a diode or not? is the new compressor, OEM or aftermarket? did it have a diode or not?

[@robertwynn3453]

a go around i have used before is to use the signal wire to power a relay and with an independant power feed .

[@Sandmansa]

I know on much older Mercedies than this one, the climate modules were troublesome. But the newer ones like you're dealing with here are noticeably better. Now if you are bypassing the switch on the compressor by supplementing a new ground through a test light and that works. But the supplied ground is not working. I would think that we may either have a poor ground going to the compressor. Or the replacement compressor may be missing a resistor. Just my thoughts. Anyway, I can't wait to see what the conclusion is.

[@needhelp2453]

It sure seems like a bad engine ground or the compressor isn't grounded too the engine properly.Try a heavy jumper wire from the body of the compressor to the negative battery terminal or to the body if you can't reach the battery.

[@sewing1243]

The first thing that should be ringing bells is that the system fails/or doesn't fail when connected to different ground points. I think that issue needs investigated first since that seems to me to indicate there are grounding issues in that vehicle.

[@A.Musa76]

Check the Control Solenoid Valve on the compressor.

[@vukasincmiljanic9777]

the diagnosDan made a video that several sensors were replaced and the car still wouldn't start, the problem was that all the sensors had the polarity reversed so the signal coming to the PCM was inverted, I know it has nothing to do with this case but what if the new the valve has reversed polarity, so where the PWM signal should be, it is actually minus, because there is a possibility that the valve receives a minus through the valve housing in addition to the minus that comes through the installation, and in that case the control wire is actually shorted to the minus because the polarity is reversed. there is also a possibility that the negative lead to the climate controller is not working. I apologize for my bad English

[@aciddiver1978]

Im so happy now i bought a 10 year old Opel mocca. Port injected, no turbo, no carbonbuildup in intake due to direct injection, no low tension pistonrings that lets oil pass, manual gear. Im gonna take good care of it, change oil frequently, because they dont make real engines anymore, just the DI crap.

[@michaellewis767]

The one thing I didn't see you check was the integrity of the ground to the controller cuz I'm assuming that it ground threw the controller and if the integrity of the ground the ground threw the controller is bad it's going to register higher amperage the only have half the ground

[@aaronkoivu]

2012 ml 350 bluetec with the same problem, customer didn’t want to go with OEM compressor but put aftermarket market, same code open or short in compressor circuit….got paid for putting the part on, now vehicle at dealership for the last three weeks with no fix, engineering/tech line involved now

[@2nickles647]

Dealership will put brand new cornpressor per tech line and charge customer tripling cost.😅

[@PineHollowAutoDiagnostics]

They will have to watch Part 2 for the top-secret Russian Hack Repair haha

[@soundman6241]

As a resident of SC, I can certainly recommend working AC for your trip to this state this time of year.😂

[@chrisanderson4799]

Maybe a bad ground between the Engine block and the AC Compressor itself? Maybe a jumper lead from the block to the Compressor to the Chassis….. Eric O,knows about crappy grounds!

[@richardrfiore]

Well I just wanted to offer some input I ain't no expert but can controller be bypassed maybe take it apart and use potentiometer in there so this way I can regulate the the current and the flow and the speed great promgram

[@TheWwong]

Love the Scotty Kilmer reference............I was thinking that ole Scotty would just say dive with the windows down!

[@petepeabody8905]

Let's not assume the junkyard part works. If the car has this issue possibly all have this fault. Ivan did you check TSB's for this issue??? Can't wait to see the fix

[@randy1ization]

dont feel bad abt the guy throwing a compressor at it.. he gambled on it in lieu of hiring a pro and he lost , that was his choice.. if u want to be a big shot and drive a big shot car, thats the price u pay... strong work..

[@russellhltn1396]

One has to stop and think about how the controller "knows" the test lights are being grounded directly though to the battery and not though the compressor. Clearly something more is going on. Either the compressor ground has a problem (causing a voltage to be applied to the controller), or it's somehow shorted to another lead the controller is monitoring.

[@x01e]

The grounding issue in the compressor is bothering me, why didnt you try to comnect a manual ground wire frim the compressor to a good ground?

[@lancecorey6582]

The ECM probably disconnects the AC when starting the vehicle to decrease the load.

[@lo1234-w9r]

The AC compressor is good and the wiring checks out fine. Replacing the AC controller didn't fix it, so that leaves a bad signal/power/ground going to the AC controller.

[@renejensen5656]

Think that 12 ohm on the control wire in the compressor sounds low, it's over 1 amp on the driver. But could be normal, would be nice to compare with the OEM compressor. The compressor is variable control type, you should be able to messure a PWM signal on the controlwire with your testlights. But if the owner has changed out the compressor, are the system been vacumed out again and the right amount of refriged put in again. Normally you check out the high and low side pressure, since there normally are pressuresensors for safety to protect the compressor. Good luck finding this issues. 🤞🤞🤞

[@Daniel-OConnell]

I don't understand why you didn't check the waveform with an oscilloscope although you had identified that it was the appropriate thing to do next. It would also have been worth checking with an alternative inductive load that drew a similar load. I suspect the key to this issue is the fact that the AC controller is managing an inductive load and the flyback diode/ suppression LC network is defective inside the control module.

[@PineHollowAutoDiagnostics]

Part 2 coming soon :)

[@jamesanderson2176]

9:00 - It's a PWM circuit, not analog, so I would definitely have the scope on that circuit. Bad ground from the compressor?

[@Garth2011]

Bad ground at the compressor ?

[@nevillegoddard4966]

Hi Ivan! I'm wondering if the compressor is not properly earthed? I'd imagine it should be earthed properly, but it's hard to imagine it not being earthed properly!

[@baxrok2.]

Thanks Ivan!

[@major__kong]

If it was working through the body ground, that's a clue. Is there something going bad with the compressor ground?

[@canadatransistor]

radiator temperature sensor may notice there is no change in temp, commands off the a/c ? Also, the ecu may notice lack of engine load based on MAF readings? The logic is probably expecting the temperature and load to rise. Likely going to a default what do i do strategy and shuts off a/c

[@georgegonzalez2476]

The compressor circuit probably has a solenoid, those have significant inductance. If the driving circuit is monitoring the startup current, it expects what an inductor does-- a linearly rising ramp going from zero amps to 1 amp over a period of some tens of milliseconds. If the compressor solenoid has even one shorted turn, that will not show up on an ohmmeter check, but the controller might notice it. Also as others have suggested, many inductors will have a reverse diode to clamp any voltage spike when the controller shuts off. If the diode is open or leaky or installed backwards, the controller may see that as a short or malfunction. But oops, there's evidence against this idea-- if the controller is okay with a light-bulb load, then it's not checking for inductance. It could still be an open or shorted diode or a noisy or flaky ground.

[@Stoney3K]

If the flyback diode was shorted, it wouldn't be able to carry 1amp of current by energizing the compressor manually, and the compressor working as intended. Instead, all of the current would flow through the diode and the compressor valve wouldn't open -- hence no cold air.

[@PineHollowAutoDiagnostics]

@@Stoney3K Very good point!

[@billsimmons7754]

On the video you never proved what the max current the controller could supply to a resistive load. You showed it could supply .8 amps. I would guess the valve is operated by a PWM signal to control the valve. I would have liked to see the scope signal to verify that. We also do not know if the control line is even supplied at a full 12 volts regardless if the control is DC or PWM. So necessary control current is still an unknown. Was that replacement compressor an OEM unit or an aftermarket rebuild? If it is an aftermarket rebuild all bets are off on the coil operating the valve having the correct resistance or reactance.

[@mrblonde2013]

Inadequate ground on the AC compressor?

[@ehsnils]

Or on the AC controller, which may explain the spontaneous behaviour.

[@andrewthompsonuk1]

Ground problem between engine and climate controller? Perhaps a slightly bad ground is fooling the current detection circuit in the controller.

[@michaelgleason4791]

I don't know for sure that you made the wrong call, but I'm really wondering why you didn't do a visual inspection and verify that there's no voltage drop from the compressor housing to the block to the battery. The controller is able to control the substituted load attached to body ground, so I would've looked elsewhere before making that call.

[@aleskyfinis1025]

I love these videos man 👨

[@mattbrown5511]

Who says that the new compressor was good out of the box. How many times have we seen new parts that were bad? I wonder if the a/c compressor has a PCB built in since it can tell if it is shorted or open circuited.

[@gerardmccarthy2432]

Hi Ivan. Love your content and help. Question will your Thinktool scanner that you are using here program an 06 range rover sport and 06 Touraeg key fobs? TIA.

[@JamesLee-oe7mt]

What happens if you run the compressor to a separate ground....will it operate at all from computer control?

[@emilpavlov147]

I Know that you probably figured it out till now. But I think the new compressor is junk.The control valve has to have a integrated super fast diode parallel connected.If not the logic think it's short circuit because of the induction created by the coil.I have seen this many times.If you put one additionaly to the circuit, everything will be OK.

[@viktorzizic8090]

You say the electric motor from the air conditioner changes speed, can you check the resistor next to it, maybe it uses the same ground

[@richardnilsen5]

9.39 of the video you said sometimes it works. Intermittent problem Scope it Tap test,tug test,pull test

[@housemaster509]

Was the compressor with the valve changed or is the valve still original? Is enough refrigerant in the system?(wrong errorcode?) What high and low pressure has the system, if valve is feeded with static 12V? Is the fan running?

[@MTLeopold]

Ah ground points. So devilishly tricky. But I might be wrong. Doesn't seem like it could be much else.

[@tedreed1201]

I had the same car and problem. It was the controller I bought on on eBay the fault code identified the controller

[@AlanBurnham]

Did I miss trying a substitute ground for the compressor?

[@mikechiodetti4482]

On to part 2.

[@Gizmo791]

Can you do a video on how to build those test lights?

[@nealcouture2]

How about on a late model E-Class the light does not come on for AC itself

[@geneirwin645]

Measure each compressor terminal to body ground (resistance) in case this compressor terminals are backwards

[@1primacron]

The junkyard one is probably junk as well lol.🤣

[@andyg5004]

Hey Ivan do Toyota exhaust manifold gasket go bad. I am in a dilemma I have to replace the donut gasket in the exhaust system. Should I also do the exhaust manifold gasket. Do exhaust manifold gasket go bad and what’s their life. I will also be replacing all the studs and bolts on the exhaust manifold bcuz when the manifold does goes bad in the future it will be easier to get out, what are your thoughts. There is no info on exhaust manifold gaskets and none of the gurus like you have covered it.

[@PineHollowAutoDiagnostics]

If it ain't broke, DON'T TOUCH IT! :D

[@jutto64]

Ground on the controller?

[@crasher88]

Possible the displacement solenoid is shorted out? Was the solenoid replaced with the new compressor or was it transfered over?

[@kwinterburn]

It's because it's looking for the return current down the black wire

[@nv1493]

I'm in the grounding issue camp 😉

[@raptervernados9630]

i can't read through all the comments nor do i know the exact circuit, but if its variable internal contolled did u check the pwm sensor that sees the internal rpm like a crank sensor basically, if the pcm isn't seeing the static rpm from that sensor it wont let it control anything. If it has that i could be thinking wrong circuit type for this.

[@georgebell3964]

Diagnose Dan had the same problem is two different compressors

[@jtjones4727]

Since that is a variable displacement compressor, is it also a clutchless design, or does it also have a clutch on it. They make both styles, not sure what MB uses. Are we sure that the a/c controller in the dash drives the solenoid/clutch if it has one? Or is it possible that something else actually provides the current to drive the solenoid/clutch such as the PCM?

[@mikesabin8568]

Depends, some are clutchless, some are not. Seen both on euro cars

[@xalty1200]

it’s clutchless but benz doesn’t use it conventionally with duty cycle. they never pulse the thing to ground you just see a b+ flatline until it shuts off for a while like how you’d expect a clutched compressor to run.

[@southhillfarm2795]

Got this shade tree mechanic hanging on the thread of curiosity.

[@rcf8367]

Jukyard contoller?

[@ferrumignis]

Did you measure the voltage on the compressor ground with respect to body ground when it was loaded?

[@PineHollowAutoDiagnostics]

If it carries current, then ground is OK :)

[@ferrumignis]

@@PineHollowAutoDiagnostics The ground could have a resistance of a few ohms which could carry some current but would raise the ground voltage.

[@zx8401ztv]

It almost sounds like a output driver transistor has semi failed. Transistors can go that way.

[@Merlyn4011]

Ouch junkyard parts on a mercedes. Sounds like the refrig solenoid is jammed

[@kens32052]

Another "Money Pit Madness" video with the GL450 lol.

[@gregmiracle9560]

Ac compressor not grounded good?

[@JoeCAlvarado]

My ac didn't work for my 2005 s430, the problem was a temperature sensor under the steering wheel.

[@michaelfraser5723]

see part 2 : }