SMC Issue on MBP 13 - Hardware Troubleshooting Guide

[cmdrdata]

However, I think the magsafe does apply 18.5V but just will not enable a higher current mode unless it gets the signal. The trickle current should be enough to power the G3Hot I would think.QUOTE]

I just have another thought on the quoted sentence above. In my previous entry I mentioned that the OPEN circuit voltage of the magsafe adapter is 6.8volt and when connected to the MBP with no G3HOT it drops down to 0.34 volt. The adapter does not output 16.5 volt until it gets the ADAPTER_SENSE signal from the SMC. The LT3470 DC/DC converter minimum input voltage is 4 volts (max is 40 volts), thus I cannot see how this G3HOT can work without battery power, initially. This is consistent with comments in another forum at macosxhints.com about "can the MBP start without battery"

In that thread from someone with an MBP with removable battery, he claimed that its battery is on the blink but not totally dead. When he starts the system and then remove the battery, the system stays on. If he then pulls the magsafe, the system died, and will not restart without the battery back in place. I think this is another indication that the battery must have at least a certain voltage level available to keep the G3HOT active at startup.

 

[Kharlitro]

well, yesterday i forgot isopropanol in lab so i couldn't clean my boards..

So i thought to use my multimeter in order to see if the dc-in board is alive or not.. and suddenly i realized that i have a multimeter but i don't know how to use it...

someone can explain me the right step to check my dc.in board?

many thanks.

 

[cmdrdata]

well, yesterday i forgot isopropanol in lab so i couldn't clean my boards..

So i thought to use my multimeter in order to see if the dc-in board is alive or not.. and suddenly i realized that i have a multimeter but i don't know how to use it...

someone can explain me the right step to check my dc.in board?

many thanks.

first of all, I doubt that the DCIN board is bad. This part has very little electronic components in it (I think just a filter capacitor and diode).
In my MBP, the DCIN board has a 5 row of solder joints that are visible when the bottom cover is removed. These solder joints (spots re-arrange the magsafe pins as follows:

Magsafe: 1. gnd 2. +v 3. sense 4. +v 5. gnd
Solder spots: 1. +v 2. +v 3. gnd 4. gnd 5. sense

So you need to find out which solder spot is 1/2/3/4 and which one is sense line or 5.
With the magsafenot connected to the MBP:
To find out which ones are gnd pins, set your multimeter to one of resistance measurement scale, ie.e, 200, 2000 or 20K, it should then display a 1 indicating open circuit. If you then touch the tip of the black and red leads together it should display a 0.x meaning that there is continuity between the red and black leads. Now touch and hold the black lead to one of the MBP screw head, and the red lead to solder spot 1-5. You should get continuity on 3 and 4. Based on this you then know which ones are 1,2 and 5.

Now connect the magsafe adapter to the MBP. Set your multimeter to the 20 volts DCV measurement. It should show 0.00 volts. As before now touch and hold the black lead to the MBP screw head. Touch the red lead probe to the DCIN solder spot 1 and 2. You should read some voltage there. If the magsafe is enabled by the SMC, you should get 16.5 volts reading. If it is not getting enabled, you will probably get a very low reading like less than 1 volt.

Any voltage reading obtained this way will indicate that the DCIN board is OK and that the magsafe is being ENABLED or DISABLED by the SMC via the sense line.

 

[Kharlitro]

first of all, I doubt that the DCIN board is bad. This part has very little electronic components in it (I think just a filter capacitor and diode).
In my MBP, the DCIN board has a 5 row of solder joints that are visible when the bottom cover is removed. These solder joints (spots re-arrange the magsafe pins as follows:

Magsafe: 1. gnd 2. +v 3. sense 4. +v 5. gnd
Solder spots: 1. +v 2. +v 3. gnd 4. gnd 5. sense

So you need to find out which solder spot is 1/2/3/4 and which one is sense line or 5.
With the magsafenot connected to the MBP:
To find out which ones are gnd pins, set your multimeter to one of resistance measurement scale, ie.e, 200, 2000 or 20K, it should then display a 1 indicating open circuit. If you then touch the tip of the black and red leads together it should display a 0.x meaning that there is continuity between the red and black leads. Now touch and hold the black lead to one of the MBP screw head, and the red lead to solder spot 1-5. You should get continuity on 3 and 4. Based on this you then know which ones are 1,2 and 5.

Now connect the magsafe adapter to the MBP. Set your multimeter to the 20 volts DCV measurement. It should show 0.00 volts. As before now touch and hold the black lead to the MBP screw head. Touch the red lead probe to the DCIN solder spot 1 and 2. You should read some voltage there. If the magsafe is enabled by the SMC, you should get 16.5 volts reading. If it is not getting enabled, you will probably get a very low reading like less than 1 volt.

Any voltage reading obtained this way will indicate that the DCIN board is OK and that the magsafe is being ENABLED or DISABLED by the SMC via the sense line.

Thank you so much! this is really what i need to know.
I've just tried to find 3 and 4 but i read resistance only in the first and the last solder pins of the DCin board..
ando so i don't understand which pin is 1 2 3 or 4.

whit magsafe connected i read a really unstable current on second and fourth pins, really unstable means that lecture change rapidly from 6.37v (max) to 0.67 (min), numbers remain on screen less than 0.5s then they change

 

[Dadioh]

However, I think the magsafe does apply 18.5V but just will not enable a higher current mode unless it gets the signal. The trickle current should be enough to power the G3Hot I would think.QUOTE]

I just have another thought on the quoted sentence above. In my previous entry I mentioned that the OPEN circuit voltage of the magsafe adapter is 6.8volt and when connected to the MBP with no G3HOT it drops down to 0.34 volt. The adapter does not output 16.5 volt until it gets the ADAPTER_SENSE signal from the SMC. The LT3470 DC/DC converter minimum input voltage is 4 volts (max is 40 volts), thus I cannot see how this G3HOT can work without battery power, initially. This is consistent with comments in another forum at macosxhints.com about "can the MBP start without battery"

In that thread from someone with an MBP with removable battery, he claimed that its battery is on the blink but not totally dead. When he starts the system and then remove the battery, the system stays on. If he then pulls the magsafe, the system died, and will not restart without the battery back in place. I think this is another indication that the battery must have at least a certain voltage level available to keep the G3HOT active at startup.

Good info. I am pretty busied out on other stuff right now but am now inclined to do a little more experimentation to understand this. I have successfully operated unibody 2.0 and 2.26 logic boards outside of the case and with no battery attached so I am just not sure how this worked. One thing I have noticed is that the non-Apple eBay magsafe chargers seem to glow green for a long time after you disconnect them. I wonder if they behave a little differently than the official Apple ones? Maybe they do not turn off the 18.5V right away after disconnecting? Just shooting from the hip since I am not near my machines to experiment.

 

[cmdrdata]

Thank you so much! this is really what i need to know.
I've just tried to find 3 and 4 but i read resistance only in the first and the last solder pins of the DCin board..
ando so i don't understand which pin is 1 2 3 or 4.

whit magsafe connected i read a really unstable current on second and fourth pins, really unstable means that lecture change rapidly from 6.37v (max) to 0.67 (min), numbers remain on screen less than 0.5s then they change

Sorry I mislead you. On my Jun 12, 2011, 05:13 PM posting I attached a picture of the bottom of the DCIN board annotated with an arrow that says magsafe pins. At that point 1 and 5 is still Gnd, 2 and 4 is +DC and the center pin 3 is still sense pin. Only at the actual logic board ]itself the pins are reordered as mentioned. Fluctuating reading at 2 and 4 means that the DCIN board is OK and the board is attempting to interface with the logic board. Since you have a liquid damaged MBP, chances are something else in the logic board is shorted or has burned out. Check the fuses as shown elsewhere in this thread. On fuses, when you put your multimeter probes on the ends of the fuse you should get conitnuity. Do not confuse the fuse with the high frequency cap (looked similar).

 

[cmdrdata]

Good info. I am pretty busied out on other stuff right now but am now inclined to do a little more experimentation to understand this. I have successfully operated unibody 2.0 and 2.26 logic boards outside of the case and with no battery attached so I am just not sure how this worked. One thing I have noticed is that the non-Apple eBay magsafe chargers seem to glow green for a long time after you disconnect them. I wonder if they behave a little differently than the official Apple ones? Maybe they do not turn off the 18.5V right away after disconnecting? Just shooting from the hip since I am not near my machines to experiment.

Hemh, curious indeed. The 2.26 is definitely the MBP SD/FW version (mid-2009). All 3 of my adapters are Apple brand, so I have no comparison to use. My thinking is that if the adapter startout with 16.5 (or 18.5 and it stays up for awhile for the G3HOT and SMC to come up, the system will power up.

 

[Kharlitro]

So bad news to me

but according to dadioh post i measured there whit magsafe in and i read only 10.85v (stable).

can i still hope in a damaged dcin board?

I am not sure what solder spots you are measuring. However, look at my attached image. You should be able to read 18.5V at the point noted. If you do not have 18.5V there then I think your magsafe board needs to be replaced (or cleaned if it is liquid damaged).

 

[peislander]

clean first

So bad news to me

but according to dadioh post i measured there whit magsafe in and i read only 10.85v (stable).

can i still hope in a damaged dcin board?

You need to clean your boards up.From the pictures i seen you should never have applied power with that much corrosion.
I've got a few back after literally bathing the boards in alcohol & drying with compressor.Good luck

 

[cmdrdata]

So bad news to me

but according to dadioh post i measured there whit magsafe in and i read only 10.85v (stable).

can i still hope in a damaged dcin board?

I'd gues that the 10.85v you measured is the BATTERY voltage. The other person that posted just recently (and dadioh) said to clean that board really good, and you should also remove the battery. If I were you, I'd do all that before doing any powering of the system. I still dount that the DC_IN board is bad. Clean that board too. It should not look dirty/cruddy like that because even if it works now, it will turn bad when corrosion would damage the PC traces later..

So bad news to me

but according to dadioh post i measured there whit magsafe in and i read only 10.85v (stable).

can i still hope in a damaged dcin board?

@Kharlitro: IGNORE my previous comments and yes, you may have a damaged DCIN board. I re-looked at the pic file you uploaded of the DCIN board,and it looked terrible. It seeemd that the damage was done quite a long time ago and the green stuff you see is corrosion which could possibly have eaten through the copper trace causing the charging problem. I would carefully take everything apart and clean everything around the spill area, and then let it dry for a a long time to ensure drynesss before powering it up again.

@Kharlitro: IGNORE my previous comments and yes, you may have a damaged DCIN board. I re-looked at the pic file you uploaded of the DCIN board,and it looked terrible. It seeemd that the damage was done quite a long time ago and the green stuff you see is corrosion which could possibly have eaten through the copper trace causing the charging problem. I would carefully take everything apart and clean everything around the spill area, and then let it dry for a a long time to ensure drynesss before powering it up again.

Actually that picture is of the corner in the logic board, so that area needs to be cleaned too. I don't know if the DCIN board looked like that, with green stuff all over. Green stuff you see is copper trace corrosion.

 

[Kharlitro]

I've just cleaned up my board, here some pics: http://img848.imageshack.us/slideshow/webplayer.php?id=dsc08376.jpg

i've also removed heatsink and replaced the ,terribly placed, original thermal grease,

now i'll wait some time before reassemble.. i'm very scared..

1 question: in that pic http://imageshack.us/photo/my-images/713/dsc08376cerchio.jpg/ you can see a chip that is under the left speaker, it looks bent and i cannot understand why..

 

[peislander]

I've just cleaned up my board, here some pics: http://img848.imageshack.us/slideshow/webplayer.php?id=dsc08376.jpg

i've also removed heatsink and replaced the ,terribly placed, original thermal grease,

now i'll wait some time before reassemble.. i'm very scared..

1 question: in that pic http://imageshack.us/photo/my-images/713/dsc08376cerchio.jpg/ you can see a chip that is under the left speaker, it looks bent and i cannot understand why..

Hey,i believe that's just a plastic cap on top of chip.Take a look with magnifying glass,you may be able to put back in place but it probally won't be an issue as it is.
Good luck

 

[cmdrdata]

I've just cleaned up my board, here some pics: http://img848.imageshack.us/slideshow/webplayer.php?id=dsc08376.jpg

i've also removed heatsink and replaced the ,terribly placed, original thermal grease,

now i'll wait some time before reassemble.. i'm very scared..

1 question: in that pic http://imageshack.us/photo/my-images/713/dsc08376cerchio.jpg/ you can see a chip that is under the left speaker, it looks bent and i cannot understand why..

Definitely looked much better. I would also looked with a magnifying glass at the PC (printed circuit) traces around the DC_IN area in the logic board to see if they are intact since these are the lines exposed to whatever liquid was spilled on it. Also on the DCIN board itself. Next to the DC_IN board plug on the logic board there is that little black and long thing. I believe under that item is the DC_IN fuse (a white rectangular stuff). Check that fuse to see if you get continuity. If that fuse is open circuit, you will not be able to power the system from the adapter and cannot power the unit from it.

 

[peislander]

Macbook pro & unibody power problems

Hey All
I'm very interested in trying to figure this charging/power problem.I've got a few Macbooks that need fixing.I can use a meter just not sure where to test & what readings i'm looking for.Let me know if you need any measurements & i'll be glad to help.Here's some of the laptops for fixing.

2008 a1286 15 inch macbook pro....magsafe don't light green & won't charge,Won't power up anymore,it did & all worked but fans went fast.
2009 a1278 13 inch macbook pro...magsafe don't light green but will very very very slowly charge, Everthing else works
2008 a1278 13 inch macbook...magsafe don't light green & won't run on power cord but everything works with charged battery
2008 a1278 13 inch macbook...magsafe don't light green & it won't power up with cord but put a battery in & it works fine & will very very slowly charge battery
2008 a1278 13 inch macbook...magsafe lights green but won't charge or power up on battery,works great when plugged in.unplug & it will shut off or go to sleep unless you hold any keyboard key & it will stay going
2008 a1278 13 inch macbook...magsafe don't light green & fan just runs full speed with only a battery hooked to it & a 470 ohm resistor gets real hot.Its the resister just down from the lcd connector.

Thanks for any help or ideas

 

[cmdrdata]

Hey All
I'm very interested in trying to figure this charging/power problem.I've got a few Macbooks that need fixing.I can use a meter just not sure where to test & what readings i'm looking for.Let me know if you need any measurements & i'll be glad to help.Here's some of the laptops for fixing.

2008 a1286 15 inch macbook pro....magsafe don't light green & won't charge,Won't power up anymore,it did & all worked but fans went fast.
2009 a1278 13 inch macbook pro...magsafe don't light green but will very very very slowly charge, Everthing else works
2008 a1278 13 inch macbook...magsafe don't light green & won't run on power cord but everything works with charged battery
2008 a1278 13 inch macbook...magsafe don't light green & it won't power up with cord but put a battery in & it works fine & will very very slowly charge battery
2008 a1278 13 inch macbook...magsafe lights green but won't charge or power up on battery,works great when plugged in.unplug & it will shut off or go to sleep unless you hold any keyboard key & it will stay going
2008 a1278 13 inch macbook...magsafe don't light green & fan just runs full speed with only a battery hooked to it & a 470 ohm resistor gets real hot.Its the resister just down from the lcd connector.

Thanks for any help or ideas

On the ones that doesn't work without AC power (not charging either), check the voltage at the IC that Dadioh indicated in the picture he attached (looking for the 18.5v. This is the voltage from the magsafe adapter through the DCIN fuse.

On the ones that are slowly charging, I'd look at the DCIN board pins (5 solder joints in a row), and while it is charging check the voltage between pins 1 and 2 or 5 and 4, you should see the nominal adapter voltage of 16.5v-18.5v if the magsafe is enabled by the SMC (IC part number 2117).
Most of the models you indicated have removable battery, so I'd try a KNOWN good battery and try that charging circuit to see if they work normally with agood battery.

 

[cmdrdata]

An update on my MBP battery and system:

My MBP (mid-2009) has an internal battery that died, and this happened to many people it seemed by googling the issue. Usually it exhibited this problem by random shutdown and low battery life, or reporting weird battery status, eventually it won't power up the system at all. What happened here is that the battery safety circuit (inside the pack) shuts down the battery output and thus no longer able to power the system, including keeping alive the "G3HOT" power supply that is supposed to be up 24/7. Today, I hooked up a 10kOhm resistor between pin5 (center pin) of the battery and its ground pin. This tells the battery to enable the DC output from the battery. Since my battery is totally depleted, I measured only a fraction of a volt between GND and +V. See the attached picture.

WARNING: before you do what I am about to do next, know how to charge a Li-Ion battery manually if you want to try it yourself!!!! Doing it wrong could cause explosion and fire!!!!!!!!!

I then hooked up a spare PC laptop power supply (output = 16.5 VDC) to the GND and +V of the battery pack and a voltmeter to the same terminals. I immediately saw the voltage rise up to 10.65v and gradually goes up to 11.9v (in about 45 minutes). I then remove the +V supply and trickle charge teh Li-Ion thru a 47Ohm resistor for about 2 hours, and I got to 12.46v. At that point I remove the charging wiring.

I put the battery back in my MBP, and the system powers up normally as I had hoped it will. Looking at the system profiler, it says that I have about about 3000mA and full capacity of 4987mA. That means that my battery is still OK. I then connected the magsafe adapter. No Green/Orange LED, totally dark, and NOT charging. The profiler says the same thing. So at this point I think I need to take the logic board out and check the DCIN fuse on the inner side of the logic board. Evidently, external DC power is not making it through the charging/power system. So bottom line is I need to trace the power logic from the DCIN board all the way to why the SMC is not enabling the adapter. BUt I made progress.

 

[Kharlitro]

i make (little) progress too,

Yesterday i reassemble the logic and tried, no charging (no green/amber, just dark) like cmrdata system profiler say no link to magsafe.
BUT, keyboard backlit that before was no working now works fine.

I read continuity in both the fuses mentioned: see pics



Uploaded with ImageShack.us

there is a method for being certain that the damage is on the DCin board?

 

[peislander]

i make (little) progress too,

Yesterday i reassemble the logic and tried, no charging (no green/amber, just dark) like cmrdata system profiler say no link to magsafe.
BUT, keyboard backlit that before was no working now works fine.

I read continuity in both the fuses mentioned: see pics
[url=http://img21.imageshack.us/img21/7194/dsc08376fusibile.th.jpg]Image[/URL]
[url=http://img829.imageshack.us/img829/4575/dsc08383cerchio.th.jpg]Image[/URL]

Uploaded with ImageShack.us

there is a method for being certain that the damage is on the DCin board?

hey,i just re-read the post & Dadioh had mentioned this ...
" the other fuse is on the underside of the board near the magsafe. That is the main fuse on the 18.5V input from the magsafe.
If that fuse doesn't have 18.5V at its input then you probably have a bad magsafe board. Give it a visual inspection for damage"
So i'm guessing you need to check at that fuse for line voltage(16.5 or 18.5v) or check the IC he mentioned with battery not connected.If you get line voltage at IC OR fuse,then dc board is good.With all the macbooks i've got i only ended up with one bad dc in board & it wouldn't send any power to the fuse.
Hope this helps
Good luck

 

[peislander]

mosfets

Ok - Just jotting this down now - don't have the Circuit Diags infront of me at the moment. will check component numbers and post some photo's once I can get to a decent camera.

What happened is this. I was trying to measure the drain of the small mosfet chip which feeds the overvoltage sense circuit to see if it was operating.
I accidentally shorted it and killed the mosfet. The mosfet blew closed circuit and is now feeding the overvoltage circuit constantly.
After the little spark the magsafe's light came on and stayed on.

My new problem is that even when the charger is unplugged the smc thinks a charger is plugged in even though it cannot read it.

I suspect that the original problem may have been that the ground side of the dual mosfet chip that enables the voltage divider that then enables the power side of the mosfet may have been blown open circuit.

Al youl should really need to do is short the source to the drain of that ground side to enable the voltage divider thus enabling the power side of the dual mosfet.

the overvoltage circuit will then enable, test the voltage and enable a connection between the magsafe and the smc via sysonewire though the bilateral switch.

I might be wrong but that is how I understand the circuit.

I am 90% sure that the problem lies in the sense circuit just after the magsafe.

Just a not on magsafe chargers.

the bit that stores the information lives in the magsafe connector.
The Powerbrick senses the load it is connected to and ramps up the voltage if the load is correct.
If you cut the magsafe connector off and use a standard PSU that supplies 16.5v you will find that the light stays on when you unplug it from the mac.
this is because when you unplug a magsafe that is connected to a apple charger the charger drops it's voltage.

Hey Chris
Any luck fixing your laptop?What mosfet's were you talking about here?
"I suspect that the original problem may have been that the ground side of the dual mosfet chip that enables the voltage divider that then enables the power side of the mosfet may have been blown open circuit."
Is there any way to test them without taking off the board?
Thanks for any help or ideas.

 

[Kharlitro]

hey,i just re-read the post & Dadioh had mentioned this ...
" the other fuse is on the underside of the board near the magsafe. That is the main fuse on the 18.5V input from the magsafe.
If that fuse doesn't have 18.5V at its input then you probably have a bad magsafe board. Give it a visual inspection for damage"
So i'm guessing you need to check at that fuse for line voltage(16.5 or 18.5v) or check the IC he mentioned with battery not connected.If you get line voltage at IC OR fuse,then dc board is good.With all the macbooks i've got i only ended up with one bad dc in board & it wouldn't send any power to the fuse.
Hope this helps
Good luck

SDENG..

That means that i have to take apart again logic from macbook, and i have to give it power , outside the chassis, with only dcin board and battery connected..

Its true?

IS this the only method??

 

[peislander]

other voltages

I am not sure what solder spots you are measuring. However, look at my attached image. You should be able to read 18.5V at the point noted. If you do not have 18.5V there then I think your magsafe board needs to be replaced (or cleaned if it is liquid damaged).

Hey Dadioh
My 2009 a1278 MBP has the issues that you mentioned on first post.It powers up & work with good battery only.I use a 16.5 volt ac for this,is this correct?It will slowly charge but magsafe is very dim green.It will power up with ac & no battery if you hold power button,then connect ac,release power button,wait a couple secs & press power button,boots with fan spin fast.
I have line voltage at mosfet in pic both sides with power off.With power on i got line voltage on all pins except pin 4 its only got 6 with power on.Does this make sense?Should all pins have 16.5 volt with power on?I have the 3.42 volts on g3hot with just ac or just battery & with both.
Thanks for starting this thread

 

[cmdrdata]

i make (little) progress too,

Yesterday i reassemble the logic and tried, no charging (no green/amber, just dark) like cmrdata system profiler say no link to magsafe.
BUT, keyboard backlit that before was no working now works fine.

I read continuity in both the fuses mentioned: see pics
[url=http://img21.imageshack.us/img21/7194/dsc08376fusibile.th.jpg]Image[/URL]
[url=http://img829.imageshack.us/img829/4575/dsc08383cerchio.th.jpg]Image[/URL]

Uploaded with ImageShack.us

there is a method for being certain that the damage is on the DCin board?

If you have an oscilloscope you can check the magsafe sense line and see if you get digital pulse signals at that line. The place to look for is at the DCIN connector (pin 5) of the logic board, and the same signal on the center pad of the magsafe DCIN board. It should come from the SMC (IC part number 2117 not too far from the DCIN connector on the logic board. That is what I plan to do if my fuse is OK (I have a Tektronix oscillsocope).

 

[peislander]

If you have an oscilloscope you can check the magsafe sense line and see if you get digital pulse signals at that line. The place to look for is at the DCIN connector (pin 5) of the logic board, and the same signal on the center pad of the magsafe DCIN board. It should come from the SMC (IC part number 2117 not too far from the DCIN connector on the logic board. That is what I plan to do if my fuse is OK (I have a Tektronix oscillsocope).

Hey,i think you need 3 volts at the pin 5 you mentioned.I tested a working a1181 macbook & had 16.5v on pins 1 & 2 & 3v on pin 5.pins 3 & 4 ground.I'm guessing the newer macbooks are the same.
Hope this helps

 

[cmdrdata]

Hey,i think you need 3.42 volts at the pin 5 you mentioned.I tested a working a1181 macbook & had 16.5v on pins 1 & 2 & 3.42v on pin 5.pins 3 & 4 ground.I'm guessing the newer macbooks are the same.
Hope this helps

The sense line of the magsafe is a digital signalling line. It may have ended up in a logic high state (in this case 3.42v), but it must have some pulse code that is decoded by the magsafe adapter (a chip embedded in the magsafe plug) to turn on the LED and enable/disable the 16.5v power pins. Dadioh in an earlier post passed a link to a website that describe this pulse signal.

 

[Dadioh]

Hey Dadioh
My 2009 a1278 MBP has the issues that you mentioned on first post.It powers up & work with good battery only.I use a 16.5 volt ac for this,is this correct?It will slowly charge but magsafe is very dim green.It will power up with ac & no battery if you hold power button,then connect ac,release power button,wait a couple secs & press power button,boots with fan spin fast.

That behaviour means that the SMC is not happy about something and is preventing the board from booting. The procedure we are describing bypasses the SMC checks and allows the board to try and boot.
So if I understand correctly your Macbook works fine on battery only. If you try to run with magsafe connected it does not work?

I have line voltage at mosfet in pic both sides with power off.With power on i got line voltage on all pins except pin 4 its only got 6 with power on.Does this make sense?Should all pins have 16.5 volt with power on?I have the 3.42 volts on g3hot with just ac or just battery & with both.
Thanks for starting this thread

I think pin 4 is the gate of the MOSFET so it may not be at same voltage as source and drain. But if you see voltage then it is passing through OK and that means your DC board is OK.

Did you have a liquid spill that brought this behaviour on? If so you need to look for signs of corrosion that may be causing this behaviour.