Determining SMD Capacitance

Thread Starter

izzy2004

Joined Mar 8, 2020
8
Hey guys,

Long time lurker finally made a account Horray!
Had a question hopefully I picked the right sub-forum for this: how would I determine the capacitance of these two surface mount capacitors near the spi flash in the attached picture? Essentially while replacing the eeprom late at night I accidentally took them off basically melted them. Replacing isnt hard however I am at a loss on what to replace them with since there are no schematics available publicly (this is from a nvidia video card). IMG_20200222_015519.jpg

Thanks for the help.

Izzy.
 

jpanhalt

Joined Jan 18, 2008
11,087
If they are caps from a power pin to ground, it is likely they are for "decoupling." The usual size is 0.1 uF; 1.0 uF is often used too, but usually in parallel with a 0.1 uF cap. If they are not power to ground, they could be anything. Presumably, you cannot test them, so I would try something in the same range as decoupling caps and see if it works.

The top one (yellow/brown) is pretty obviously a capacitor. Are you sure the bottom one (black) is a cap and not a resistor?
 

ci139

Joined Jul 11, 2016
1,898
the black ones (without markings) usually are inductors or RF supressors https://electronics.stackexchange.com/questions/338625/mlcc-capacitor-color (← i ve usoldered and measured a loads of them -- there is a reason to believe the color ident is mostly impossible -- likely nor by the size cos i ve seen really small 0.1µF ones and compared to them huges ones with similar capacity but likely tolerating higher voltages . . . https://www.quora.com/Are-these-things-capacitors/answer/Steven-J-Greenfield)
 
Last edited:

Thread Starter

izzy2004

Joined Mar 8, 2020
8
If they are caps from a power pin to ground, it is likely they are for "decoupling." The usual size is 0.1 uF; 1.0 uF is often used too, but usually in parallel with a 0.1 uF cap. If they are not power to ground, they could be anything. Presumably, you cannot test them, so I would try something in the same range as decoupling caps and see if it works.

The top one (yellow/brown) is pretty obviously a capacitor. Are you sure the bottom one (black) is a cap and not a resistor?
Thanks for the response. I am not entirely sure I just assumed they were both caps. Unfortunately at the moment all I have to work with is that picture up there that I took before the 'incident'. Problem is really getting the right values.
 

Thread Starter

izzy2004

Joined Mar 8, 2020
8
the black ones (without markings) usually are inductors or RF supressors https://electronics.stackexchange.com/questions/338625/mlcc-capacitor-color (← i ve usoldered and measured a loads of them -- there is a reason to believe the color ident is mostly impossible -- likely nor by the size cos i ve seen really small 0.1µF ones and compared to them huges ones with similar capacity but likely tolerating higher voltages . . . https://www.quora.com/Are-these-things-capacitors/answer/Steven-J-Greenfield)
Thank you Salem ill try 0.1myuF and see how it goes and try to find a similar inductor. Ill report back.
 

Thread Starter

izzy2004

Joined Mar 8, 2020
8
Assuming the Winbond 25040 EEPROM is the same as Atmel 25040 (http://ww1.microchip.com/downloads/en/devicedoc/doc0606.pdf), pin 8 is Vcc, and pin 7 is !Hold. A small ceramic cap to ground from Vcc is pretty common.

Pin 1, however, is !CS. From my experience, it would be unlikely to put a cap on that. In some instances, however, a weak pull up is placed on !CS. If a resistor, it may still be good.
I actually have a lovely diagram right here from the datasheet and well as the sheet itself if it is of any help. Thanks for the help.Screenshot_20200308-152546.jpg
 

Attachments

jpanhalt

Joined Jan 18, 2008
11,087
That's exactly the same pinout as the Atmel chip with the same number. Doesn't change my comment at all.

Here's a Google search on using a pull-up om the !CS (chip select) pin: https://www.google.com/search?newwi...hUKEwiK_Lno-YvoAhU3JzQIHcWPA1AQ4dUDCAg&uact=5

I have never found that to be necessary, even with Mode 3. However, it is not inconceivable that a manufacturer that has to comply with a variety of devices wouldn't do it.

Edit: If you have a multimeter, measure the resistance from the end of that "capacitor" (the black one) furthest from the EEPROM pin to Vcc and GND. What do you find?
 
Last edited:

Thread Starter

izzy2004

Joined Mar 8, 2020
8
That's exactly the same pinout as the Atmel chip with the same number. Doesn't change my comment at all.

Here's a Google search on using a pull-up om the !CS (chip select) pin: https://www.google.com/search?newwi...hUKEwiK_Lno-YvoAhU3JzQIHcWPA1AQ4dUDCAg&uact=5

I have never found that to be necessary, even with Mode 3. However, it is not inconceivable that a manufacturer that has to comply with a variety of devices wouldn't do it.

Edit: If you have a multimeter, measure the resistance from the end of that "capacitor" (the black one) furthest from the EEPROM pin to Vcc and GND. What do you find?
Hi there. I am not able to measure it as I have melted off the two components and am trying to determine the values based off the one picture I took. But thank you for your help. From your link and the help of the others who replied I determined that the black thing is a resistor valued at approx 10kohm and the brown bugger is a cap at 1myuF so I will try that this evening and report on how it went.

Hopefully the thing finally reads. That was the main issue basically. My programmer writes just fine however the the eeprom is never detected by the video card. Thanks guys.
 

Janis59

Joined Aug 21, 2017
1,834
Why not to make any direct measurement?? The voltage in modern-day capacitance meters is beyond the 60 mV thus no any p-n barrier may not be opened, thus there is no need for desoldering a component before very measurement.
 

Thread Starter

izzy2004

Joined Mar 8, 2020
8
Why not to make any direct measurement?? The voltage in modern-day capacitance meters is beyond the 60 mV thus no any p-n barrier may not be opened, thus there is no need for desoldering a component before very measurement.
What are you suggesting I measure? What would I be looking for? I would just get some random resistance between the two pins of the eeprom. If you are refering to the two electrical components maybe you missed what i wrote about me melting them off accidentally while using a decent amount of solder desoldering the eeprom. Was afraid of ruining any electrical traces.
 

Thread Starter

izzy2004

Joined Mar 8, 2020
8
That's exactly the same pinout as the Atmel chip with the same number. Doesn't change my comment at all.

Here's a Google search on using a pull-up om the !CS (chip select) pin: https://www.google.com/search?newwi...hUKEwiK_Lno-YvoAhU3JzQIHcWPA1AQ4dUDCAg&uact=5

I have never found that to be necessary, even with Mode 3. However, it is not inconceivable that a manufacturer that has to comply with a variety of devices wouldn't do it.

Edit: If you have a multimeter, measure the resistance from the end of that "capacitor" (the black one) furthest from the EEPROM pin to Vcc and GND. What do you find?
I also am not familiar with what mode 3 means if you could elaborate?
 

jpanhalt

Joined Jan 18, 2008
11,087
I also am not familiar with what mode 3 means if you could elaborate?
Are you familiar with SPI?

There are two settings for the clock. All devices require setting clock polarity (i.e., is idle state high or low). There is also a phase/edge setting that describes when during a clock cycle data are sent. The edge setting is used by most PIC and ARM-based controllers. Others use phase. Edge is the inverse of phase. Regardless of nomenclature, two settings each of which can be 1 or 0 gives 4 possible combinations. Those are conveniently called Modes numbered 0,1,2, and 3, which correspond to the binary values of the polarity and phase settings, respectively. Thus, polarity = 0 and phase = 0 is defined as Mode 0. Similarly, polarity =1 and phase =0 is Mode 2, and forth. Wikipedia has an extensive discussion of SPI and its conventions.

That has nothing to do with the toasted resistor, except to help determine whether the two terminal device in question is/was a resistor or capacitor, and if a resistor, what its value might be. My assumption was that it was a weak pull-up on the chip select (CS) line as it would be unusual to attach a capacitor to that pin.
 

Janis59

Joined Aug 21, 2017
1,834
What are you suggesting I measure? What would I be looking for? I would just get some random resistance between the two pins of the eeprom. If you are refering to the two electrical components maybe you missed what i wrote about me melting them off accidentally while using a decent amount of solder desoldering the eeprom. Was afraid of ruining any electrical traces.
You asked how to know what capacitance have a specific probably badly marked capacitor, I was answering that even if it is shunted the measurement without desoldering is second worth deal.
 

Thread Starter

izzy2004

Joined Mar 8, 2020
8
Are you familiar with SPI?

There are two settings for the clock. All devices require setting clock polarity (i.e., is idle state high or low). There is also a phase/edge setting that describes when during a clock cycle data are sent. The edge setting is used by most PIC and ARM-based controllers. Others use phase. Edge is the inverse of phase. Regardless of nomenclature, two settings each of which can be 1 or 0 gives 4 possible combinations. Those are conveniently called Modes numbered 0,1,2, and 3, which correspond to the binary values of the polarity and phase settings, respectively. Thus, polarity = 0 and phase = 0 is defined as Mode 0. Similarly, polarity =1 and phase =0 is Mode 2, and forth. Wikipedia has an extensive discussion of SPI and its conventions.

That has nothing to do with the toasted resistor, except to help determine whether the two terminal device in question is/was a resistor or capacitor, and if a resistor, what its value might be. My assumption was that it was a weak pull-up on the chip select (CS) line as it would be unusual to attach a capacitor to that pin.
Interesting enough apart from flashing them and messing around with the roms they hold I do not unfortunately know how they work. I read the datasheet on a few of them but they are definitely complex devices. But its only natural. I did flunk out of my EE bachelors after all haha.

Im happy though you mentioned these modes it is interesting and something i have never heard before. The fact that it uses a pullup resistors will help. Im going to try to solder on a compatible cap and resistor in a few hours
 

N5KS

Joined Apr 17, 2016
6
Seems that the upper component is likely a bypass cap which shows that /HOLD pin being held high on the 25Q40. Most likely values I’ve experienced is 100nF or 10nF. The bottom component appears to be a resistor used as a pulldown for the /WP (write protect) pin, as it bridges gnd and /WP. Hard to say the value as I’ve seen 1k to 100k.
 
Top