Help reading schematic voltages input/output

[gjgib]

Hello, this might be embarassing since I consider it a dumb question but keeps me kinda stuck sometimes in my journey.

I want to know if there is any post in this forum which explains my questions. A) When, where and how do I know when X voltage line is input or output. Example:


I'm battling with this IC to understand it's voltages. It's from non-turning-on Lenovo Ideapad 3 15iil05 - NM-D461.

I see some lines like +5VL / +5VALW / V9B+ / +3VALW being okay and some others like 1.8VALW / +0.6VS / +2.5V_DDR. I just know that some of them are bad or missing, but I don't know entirely which way they are going, if IN or OUT. Btw: +V9B has its 20v.

B)Any methos to know where values going for sure? Besides logic...

I bought a couple of these LV5116AGQW IC's time ago and I'm ready to replace but first I must be 100% sure.

I was following lines and realized that +VCCIN is missing right here:


At this point I'm kinda stuck because since I'm rookie there's a bunch of voltages I don't know their usual value.

This IC, which I think is responsible of sending some of these voltages to the mosfets has these 2 values.


If there's posts already with the info I'm requiring feel free to just leave the links here to avoid taking other's valuable time.

Thanks for any help and sry for still being noob hah Xd

[mcplslg123]

You can download the datasheet of IC,if available, to understand it better. +VCCIN will come much later , so not of concern yet.
1.2V(ddr power supply)/2.5V both are associated with ram circuit. Unless you have 1.2V/2.5V/1.8V , cpu core power related supplies wont rise.
So in short, focus on why you're missing 1.2/2.5/1.8V rails. Check the respective enables for these rails.
IC will output these rails when
01.All its inputs are present( means its powered)
02.All its enables are present
03.There is no short on output rails.

Hope this helps you in focussing a bit more sharply.

[gjgib]

I thought +VCCIN was supposed to come online early as after the +V9B... Jesus, I just have no idea. lol. Okay so I'm going again with PU9800 (LV5116AGQW) voltages and enables as you said.

Thing was I also guided me from the power block and thought brackets meant to be in this order:


So I thought it was first block 1, then block 2 and block 3 and, like that, supposedly +VCCIN was before 1.2V/2.5V/1.8V. Which is wrong then...

Thanks so much for clarifying. Now I know better.


edit1: About the voltages in and out. Any thoughts?

[gjgib]

edit1: About the voltages in and out. Any thoughts?

PUMP

[mon2]

The high adapter voltage feeds into the lower voltage regulators but the EN (enable) decide when that regulator is to turn ON / OFF.

[gjgib]

The high adapter voltage feeds into the lower voltage regulators but the EN (enable) decide when that regulator is to turn ON / OFF.

Thanks again mon2 but I meant about how I know when voltages are going in and out in a schematic. Like in the first 2 pictures I shared. The voltages that come from top of the IC are going in and the ones at the sides are going out? Or is that view wrong?

edit1: there's also some colors (black, red) in some voltages letters/numbers. They mean something?

[mon2]

A good library part drawing of the charger component should offer labels that detail the vitals like the power rail.

VCC ; used to power the component.
VSS / GND ; ground for the component.

Review the drawing and their pin numbers - the label has _IN suffix so that is an INPUT to this IC.

Respectively, if there is an _OUT = output from the IC.

If there is a '#' suffix, then this is considered to be an ACTIVE LOW signal. Meaning that if the signal is a logic LOW = ~0 volts, then the signal is TRUE or ACTIVE. If the signal is HIGH = ~3v3 (usually), then the signal is NOT ACTIVE. This is based on if the IC is powered @ VCC by 3v3. These days we are designing with parts that have 0v9 or 1v2 or 1v8 for the VCC - these rails are very common inside of the new macbooks / iphones, etc. Respectively, their input and output voltage swings must be kept within safe ranges else can damage the component.

Some pins are 'tolerant' to higher than the VCC voltage that is used to power the IC and this is through internal clamping diodes. The internal diodes will shunt or keep the voltage to a pre-defined safe range.

For example, if an IC is operating @ 1v8 for VCC but outside signal is 3v3 - the 2 parts cannot mate with each other safely unless there are level shifters / voltage translators (same thing; different words) to shift from one voltage to the other. The other method is to purchase the same part which is 'tolerant' to the higher voltage without risk of damage.


Recommended reading:

Getting_Started_in_Electronics_-_3ed_-_[Forrest_M.Mims].pdf (zpag.net)

[gjgib]

A good library part drawing of the charger component should offer labels that detail the vitals like the power rail.

VCC ; used to power the component.
VSS / GND ; ground for the component.

Review the drawing and their pin numbers - the label has _IN suffix so that is an INPUT to this IC.

Respectively, if there is an _OUT = output from the IC.

If there is a '#' suffix, then this is considered to be an ACTIVE LOW signal. Meaning that if the signal is a logic LOW = ~0 volts, then the signal is TRUE or ACTIVE. If the signal is HIGH = ~3v3 (usually), then the signal is NOT ACTIVE. This is based on if the IC is powered @ VCC by 3v3. These days we are designing with parts that have 0v9 or 1v2 or 1v8 for the VCC - these rails are very common inside of the new macbooks / iphones, etc. Respectively, their input and output voltage swings must be kept within safe ranges else can damage the component.

Some pins are 'tolerant' to higher than the VCC voltage that is used to power the IC and this is through internal clamping diodes. The internal diodes will shunt or keep the voltage to a pre-defined safe range.

For example, if an IC is operating @ 1v8 for VCC but outside signal is 3v3 - the 2 parts cannot mate with each other safely unless there are level shifters / voltage translators (same thing; different words) to shift from one voltage to the other. The other method is to purchase the same part which is 'tolerant' to the higher voltage without risk of damage.


Recommended reading:

Getting_Started_in_Electronics_-_3ed_-_[Forrest_M.Mims].pdf (zpag.net)

Very insightful, mon2. Thanks a lot. I'll go through that guide. Very much needed.

I'll also proceed to measure and list all the voltages on this IC and post it here in case I miss whats going on since I can't find anywhere the datasheet for this specific 40 pins IC (LV5116AGQW). The one that comes in is a 20 pins IC named LM5116WG.

[mon2]

The datasheet is another world in itself. The mainstream companies are often in the USA (or at least were since the late 60s) - they invented the silicon and overseas companies copied the IP. We called it 'outsourcing' but I call it creating a Frankenstein. The servent killed the master scenario. There are some really good documentaries on the birth of Silicon Valley - worth a watch and inspiring. Years ago, met with Steve Wozniak (Apple), Tramiel family (Commodore / Atari) and founders of many other tech companies. Very humble folks and Steve Wozniak had body guards at the time. He was showing off the industry's first universal remote control at the time.

The datasheets can vary with the vendor who builds the product and many of them are in the witness protection program. We will not have access to each and every datasheet and yours may be one of them. Found the 20 pin version from Texas Instruments but keep hunting. Also if stuck, post on EEVBLOG for help - someone there may have what you are after.

[gjgib]

Wow, meeting those must've been inspiring. Thanks mon2 I'll look into it. Still to jump deeper into the electronic roots and any documentary is very welcome.

Btw, I managed to, carefully, measure every pin in the 40 pin LV5116AGQW. Any thoughts with these results?




VSYS - 20,2
VCC - 5,1v
PVCC - 5,1v


VDDQ_HG - 0v ----------------V1P8A_IN - 3,3v
VDDQ_BOOT - 4,6v ----------V1P8A_PH1 - 0,268v
VDDQ_PH - 0v -----------------V1P8A_PH2 - 0,268v
VDDQ_SWSNS - 0v -----------V1P8A_SNS - 0,268v
VDDQ_LG - 0v -------------------V1P8A_CTRL - 5,1v
VDDQ_SNS - 0v
VDDQ_CS - 0v

VPP_IN - 3,3v
VPP - 0v

VTT_IN - 0v
VTT - 0v
VTT_SNS - 0v -----------------------SCL - 3,3v
-------------------------------------------SDA - 3,3v

AUX_PWM - 0v ---------------------PROCHOT# - 0,39v
AUX_SWSNS - 0v ------------------PG_ARAIL - 0v
AUX_SNS - 0v -----------------------PG_DDR - 0v
AUX_VID0 - 3,3v --------------------DDR_ID - 0v
AUX_VID1 - 3,3v
AUX_CS - 0,25v ---------------------PMIC_EN - 3,3v
AUX_SET - 1,2v ---------------------DIGITAL_CTRL - 5,1v?
AUX_RGND - 0v --------------------SLP_SUS# - 3,3v
--------------------------------------------SLP_S4# - 3,3v
--------------------------------------------DDR_VTT_CTRL - 3,3v


(--- just to save space)

[gjgib]

Pump. Did I ask for wrong stuff?

[mon2]

Remove all power. Meter in resistance mode.

Measure the resistance (not diode mode) of pin # 25 to ground. You can measure at the inductor / coil on the same pin. This is to check if this line is shorting to ground through some component that is defective or leaky. The concern here is that this rail should be supplying a +1v8 voltage rail but is not.

[gjgib]

Thanks for reply mon2. There's 3.4KΩ +pin25 -GRD. And 3.8KΩ -pin25 +GRD. (+- probes).

Is that value enough to consider there's a leak in the +1v8?

[mon2]

For resistance mode tests, there is no polarity requirement with the meter probes.
For diode mode tests, the red meter probe must be to ground and black to the point to test.

At ~3k4 = 3,400 ohms to ground, this measured rail is not shorted.

[gjgib]

mon2 Thanks for insight. Is there a value treeshold for a line to be considered 'shorted to ground'? I know it depends on some other things, but I meant this the basic way.

Btw: Should I dig more into a possible fault on the IC inputs or do you think, considering the values, that is enough evidence for the IC to be considered as faulty? I know and understand that 100% of the responsability of this is on me but the rewards can be shared

[mon2]

In resistance mode - in my opinion, if the measurement if say 500 ohms or lower, it is of concern. Some engineers like to place a 100-120 ohms to ground resistor intentionally on the power rail and this part then acts like a bleed resistor. When the power is OFF, the resistor will impose a relative heavier load to cause the current to bleed off and discharge the capacitors to zero volts. In summary, when there is a load present, the voltage will dissipate from the capacitors that are holding the same. The capacitors will discharge and this makes it safer for the next power on cycle to start from zero.

Study the other package for this part to review how this part operates. If available, check if there is an enable pin for this power rail. Check if that enable signal is present. It does sound like this power rail is defective and the IC should be replaced. Before proceeding, check the rest of the output power rails on this component. If their voltage is not present, perform the same type of testing -> is the output shorted? If shorted, the short condition must be resolved before applying a replacement part.

When you have time - this is a must watch. I think we should have a sticky here for such worthy documentaries. I know there is another excellent one on the birth of the semiconductor industry - on how Fairchild Semiconductor and others were started. I will try to find that one as well. Fairchild is now owned by Texas Instruments.

Triumph of the Nerds (1 of 3) 4K Upscaled (youtube.com)
* must watch documentary

[gjgib]

Thanks for the documentary mon2 . Already seems like a valuable life lesson. How the dudes following their passion kinda self privated from other life pleasures changed the world.

I detected low impedance in the +VCCIN rail (68,6Ω). +VCCIN_AUX (174Ω). And +VDDQ_OUT (+1.2V)-(418Ω) Started doing some tests. Would you consider this amount okay for core rails nowadays or somethings off?

[mon2]

+VCCIN rail (68,6Ω)

Yes, this is considered to be a low resistance.

Remove all power. Meter in resistance mode. Review the resistance across the mosfet pins for this power rail (+VCCIN).

Check:

source (1-2-3) & drain (5-6-7-8)
source (1-2-3) & gate (4)
gate (4) & drain (5-6-7-8)

Perform the above checks for the high side (PQ3501) and then the low side (PQ3502) mosfets for this rail. Post all measurements. On a normal mosfet, the resistance should be high.

[piernov]

68 ohms is fine for Ice Lake/Tiger Lake CPU main power rail.

[gjgib]

mon2 Found something fishy. Might be the issue(?

PQ3501 Impedance / S-D 0.6MΩ / S-G 2.8MΩ / G-D 12MΩ

PQ3502 Impedance / S-D 88Ω / S-G 2.4MΩ / G-D 5.4MΩ

Alsoo... In Diode Mode:

PQ3501 (SD 0.5V) (DS 2.4V) (GS 1.07V) (SG 0.14V) (GD 1.35V) (DG 2.5V)

PQ3502 (SD 0.18V) (DS 0.05V) (GS 1.17V) (SG 0.45V) (GD 1.25V) (DG 0.5V)

Second one seems like missbehaving right? Should I remove it for testing outside board?



piernov Thanks for the insight, I've created a .txt file to save all about rails resistance values based on generations. I feel like its useful to avoid confusions like this in the future.