MCU reset line shorted low, can I change pull up resistor?

[caphair]

This is more of a general question to something I've seen more than once.

If a MCU or CPU etc reset line is pulled low due to the MCU, could you change the pull up resistor to a lower value (for example 100k to 1k) to pull the line up to the proper voltage to get it going or is that a bad idea?

Also what would happen if the resistor was replaced by a wire? Too much current going to ground or?

[redwire]

Re: MCU reset line shorted low, can I change pull up resistor?

I would say you can try force Reset high using a low value pullup resistor, say 100R-1k. It will stress the IC keeping the line low. If you think the line is shorted, ohms vs diode-test should show that.

What you might be seeing is an (external) watchdog tripping and holding the MCU reset line low.
If an MCU cannot boot up and run code, it will not be able to toggle the watchdog timer, and after a delay WDT pulls reset low.
Or, the rail monitor can trip if VDD is low say 2.8V instead of 3.3V and it holds reset low.

Usually there is a factory test-pad/jumper to over-ride the watchdog timer, so the factory can initially program a blank MCU. A blank MCU would not reset the watchdog and it would trip before flash memory was programmed.

A typical watchdog timer IC TPS3823

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

What would be a watchdog timer? I have here an iPhone with reset pulled low. Should be 1.8v when you power the device on.

Diode mode shows right value of .360v compared to a known good board. I just figured the CPU was bad.

I tried a low 10ohm resistor (only smd resistor I had compared to original 100k) now reset line is high at 1.8v but still doesn't power on

[redwire]

Re: MCU reset line shorted low, can I change pull up resistor?

I think you are working on iPhone 6S from post here?
I know almost nothing about its H/W operation. I can guess and maybe something to try.

Dialog Semiconductor D2255A is custom power management IC. They might send a datasheet if you are Samsung, Apple, Nokia...

It offers a few reset functions.
Three reset inputs -
pin P7 is watchdog input from AP
pin P8 is TRISTAR reset command
pin P9 is SOC reset

This and more make the output pin K4 PMU_TO_SYSTEM_COLD_RESET_L
If this pin is stuck low, look at the other three reset inputs. I think they eventually get OR function.
Or maybe the PMIC is in shutdown (pin N8) mode.

It looks like SYSTEM_ALIVE might be another watchdog pulse output going to other sections.

It sound like touch disease and screws shorting to traces are problems on these

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

I think you are working on iPhone 6S from post here?
I know almost nothing about its H/W operation. I can guess and maybe something to try.

Dialog Semiconductor D2255A is custom power management IC. They might send a datasheet if you are Samsung, Apple, Nokia...

It offers a few reset functions.
Three reset inputs -
pin P7 is watchdog input from AP
pin P8 is TRISTAR reset command
pin P9 is SOC reset

This and more make the output pin K4 PMU_TO_SYSTEM_COLD_RESET_L
If this pin is stuck low, look at the other three reset inputs. I think they eventually get OR function.
Or maybe the PMIC is in shutdown (pin N8) mode.

It looks like SYSTEM_ALIVE might be another watchdog pulse output going to other sections.

It sound like touch disease and screws shorting to traces are problems on these

It's actually a 6 and that schematic is for a 6s which is more complex with it's ready/request pins

The 6 I'm referring to uses just one reset line that goes to a charge management chip (Tristar), display PMU, and to the CPU maybe for the soc I'm not sure.

I tried removing both Tristar and display PMU, Reset was still low. The pmic provides the reset and it's a steady 1.8v at the pull up resistor so I assume it's fine.

When I replaced the 100k pull up with a 10ohm the reset test point showed high of 1.8v like it should be but still phone didn't boot. So I was wondering if the CPU is dead or maybe bc I tried with a too low ohm resistor it didn't work?

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

Here's the iPhone 6 reset side of the pmic

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

If I'm not measuring a short on the reset line and get same diode reading as a known good board, does that mean ALL the reset inputs are ok or only the reset output is ok?

[budm]

Re: MCU reset line shorted low, can I change pull up resistor?

If you measure in Ohm mode, what do you get?

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

If you measure in Ohm mode, what do you get?

I'll measure when I'm home. When I test in diode mode I put red probe on ground and black at the reset test point.

Is measuring in ohms more accurate?

[Longbow]

Re: MCU reset line shorted low, can I change pull up resistor?

The processor itself does not "send" or create a master reset. Be sure which reset you refer to. The main power-up reset is created by a special reset IC, or sometimes by a simple RC circuit connected to a transistor. If the reset pin is constantly held in reset, then your reset circuit is faulty (typically a shorted digital reset transistor).

There are also reset outputs from the main micro that reset other ic's in the unit. If the reset output is held low, you have to determine whether your main micro isn't ready, or if something downstream is shorted. It isn't that simple. Refer to the complete diagrams and schematics that the phone manufacturer always supplies you with.

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

The processor itself does not "send" or create a master reset. Be sure which reset you refer to. The main power-up reset is created by a special reset IC, or sometimes by a simple RC circuit connected to a transistor. If the reset pin is constantly held in reset, then your reset circuit is faulty (typically a shorted digital reset transistor).

There are also reset outputs from the main micro that reset other ic's in the unit. If the reset output is held low, you have to determine whether your main micro isn't ready, or if something downstream is shorted. It isn't that simple. Refer to the complete diagrams and schematics that the phone manufacturer always supplies you with.

So in the schematic I've attached do the 3 input resets cause the output reset of 1.8v to be present?

The reset test point is for the output reset of 1.8v and that's what's missing. There are two other ic's connected to that output reset which I've removed and the reset still holds low. In the past I've fixed this issue by removing either one of those two ic's.

This time it made no change. So does that leave the CPU as the culprit?

[redwire]

Re: MCU reset line shorted low, can I change pull up resistor?

This is not a simple reset circuit. Same principle as iPhone 5C.
I think the PMIC master reset output looks at the three reset inputs, and uvlo, and watchdog heartbeat SYS_ALIVE.
The PMIC has three reset inputs and a master reset output and watchdog heartbeat. PRE_UVLO looks like it goes low for undervoltage lockout.

The schematic mentions the pullup resistor PU or pulldown PD on the pins.
Example RESET_IN3 "100-300k INT PU to LDO12" means the PMIC has an internal pullup resistor 100k-300k value to the LDO 12.

I would look at the 5 signals going into the PMIC reset circuit.
It could be a bad PMIC or one of the three parts is holding it low.
Again, this is just guess work without the PMIC datasheets.

[Longbow]

Re: MCU reset line shorted low, can I change pull up resistor?

So does that leave the CPU as the culprit?

That would not be my first guess. The micro might just be doing its job. Since you have 3 lines marked reset IN, I would want to scope all those pins and actually SEE if you get activity on all 3 at power up. If the RESET IN pin voltages just sit there, then the micro is waiting for something to happen.

The more general question you asked is whether you might change a pullup resistor to force the circuit into operation and the answer is definitely NO. If you are thinking about changing part values, you are really saying that one of the micro outputs is partially faulty and that you can help it along, by changing resistor values.

Since it is easy to compare with another unit, just take a resistance reading from the pin to ground. That will tell you 95% if there is pcb contamination, or an obvious fault at the micro output pin. Personally, I find it odd that the control micro waits for several resets and then issues one of its own. I wonder if things are just labeled incorrectly? It makes more sense that the micro would receive 1 power up reset, then issue 3 resets to supporting chips. Think about it. The control micro has to be up before anybody else.

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

@Longbow

I checked all the input resets and the soc line which is on a 1.8v rail was shorted. I relieved the short but it didn't solve the issue. It actually made the CPU extremely hot to touch afterwards.

The soc is part of the CPU so I chalked it up to faulty CPU

[Longbow]

Re: MCU reset line shorted low, can I change pull up resistor?

What type of short did you find, pcb contamination? It doesn't make sense that fixing a short would cause something to heat up, but anything is possible I guess. Remind me please what is soc? I know this as System on Chip, but you obviously have another meaning.

[caphair]

Re: MCU reset line shorted low, can I change pull up resistor?

What type of short did you find, pcb contamination? It doesn't make sense that fixing a short would cause something to heat up, but anything is possible I guess. Remind me please what is soc? I know this as System on Chip, but you obviously have another meaning.

The short was a capacitor on the line. Not sure why didn't work after and CPU got hot. Phone had slight water damage.

Soc is system on chip, it's part of the CPU but doesn't make up entire CPU in the iPhone. Not sure exactly it's function