Tinkerboard S: What is ASUS view on voltage drops in cables?

[king.oslo]

Hello there.

 

At present I power my Tinker Board S with a bench power supply set to 5.25V. The power supply has 10A current limiter. I measure the voltage at the GPIO pins using a 6.5 digit multimeter (in cal). The micro usb cord is 1.2 meters (4 feet).

 

When the board is running in tinker os at idle, it is drawing 500mA, and i read 5.00V on the GPIO. But when I load the CPU using sysbench the current spikes to 1.6A, and the voltage measured on the GPIO drops to 4.45V. I measured that the micro USB cable is 0.37Ohm resistance, so about 0.55V of that power drop is due to the micro usb cable. That means all of the voltage drop is not due to the power cord, and the voltage onboard at the micro usb connector is exactly 4.6V, and I have some questions:

 

0. What is causing of the power drop of 0.25V measured on the GPIO pins relative to onboard at the micro usb connector? This is a cause for concern if I want to attempt to power the board using the GPIO.

1. In the thread below, TonyMac32 propose to drive the tinker board using the GPIO pins directly. Are there any other viable options?

2. I get a low voltage warning in the OS when I start loading the CPU with sysbench, and I read this thread:

Since this is not a plug and play solution to the problem, what are ASUS recommendation to deal with the voltage drop in the usb cable?

3. What is the absolute maximum voltage rating of the tinker board s? I am not asking about the USB maximum voltage specification, I am asking what the onboard hardware has as maximum rating.

 

Thank you for your time.

 

Kind regards,

Marius

[TonyMac32]

Firstly I have not tested a Tinker S specifically, the power input I  different. But the answer is no, without using gpio you cannot overcome the voltage drop. I have no read the data sheet for the RK808 PMIC, but I doubt you can cross the 5.5V threshold without possible damage.  

[tkaiser]

14 hours ago, king.oslo said:

what are ASUS recommendation to deal with the voltage drop in the usb cable?

 

ASUS is not here. You might ask them where they are (no idea) and most probably they tell you to use a PSU with higher amperage ratings since it's so much fun to use this Micro USB crap in the first place and then fool all the unfortunate users suffering from the usual voltage drop problems afterwards (at least that's what the RPi clowns do who invented the 'great' idea to use this shitty connector for DC-IN).

Edited August 6, 2018 by tkaiser
Frank Wu seems to be from ASUS

[arox]

I understand you use a bench or enterprise PSU with high amperage, voltage regulation and security ...

 

So why should you be bothered by an 1,2 m USB cable ? Cant you use another cable, short and with big wires ?

 

What I always asked myself is the resistivity of the contacts on micro USB and how much it can change from one plug to another or over time. I spent the day trying to fix my old audio amp and speakers (more than 20 years) : the potentimoeters are dead,  the resistivity change with the charge (or when I adjust them) and so the sound is horrible. Imagine what happend to your board if some crappy contact do the same when the processor demand a power burst ! The response is that the board can crash anytime but more probably when doing heavy computation or accessing disk or network ...

[TonyMac32]

I provided a quick answer via my phone earlier so here is a bit more complete:

12 hours ago, king.oslo said:

I measured that the micro USB cable is 0.37Ohm resistance

 

Use a shorter cable if at all possible, and check the conductor size.

 

12 hours ago, king.oslo said:

I load the CPU using sysbench the current spikes to 1.6A, and the voltage measured on the GPIO drops to 4.45V.

 

12 hours ago, king.oslo said:

and the voltage onboard at the micro usb connector is exactly 4.6V

 

OK, so 150 mV of loss isn't great, but it's not unheard of either.  The "S" has, as you pointed out, some hardware for detecting power supply issues, I would guess this could potentially result in some minimal losses the standard Tinker did not have.

 

My Tinker S, using a 1.5 meter USB cable I purchased on the basis of it's wire gauge (20 AWG I think?  I doubt it's 18...), 5.25 V supply, 10 Amps, is reading 5.17 Volts at idle on the USB Master port.  running "minerd --benchmark" to load the CPU I dropped to 4.8.

[Frank Wu]

 

Hi~

 

About how we suggest to prevent the voltage drop.

Not much but can reference,

1. 5V and full power as the spec. (e.g. some supply said 5v/2a (10W), but actually, when you asked v2a, you would see it may only provide 4.8v/2a)
2. We would suggest using 3A (15W) to prepare for the high loadings.
3. The cable with 20~18 AWG for large current power usage.

4. We also provide the best and official choice, the Tinker Power Supply.

 

And you can check what is the current voltage in the system.

 

Take a look with this node, "/sys/bus/iio/devices/iio:device0/in_voltage2_raw".

If it had been there and assumed the function was normal. You can refer below sample code (python) to get voltage.

DETECT_VOLTAGE = 4.65 #4.65
ADC_IN2_RAW_PATH = '/sys/bus/iio/devices/iio:device0/in_voltage2_raw'

with open(ADC_IN2_RAW_PATH) as in_voltage2_raw:
  val2_raw = int(in_voltage2_raw.readline())
  val_input = float(val2_raw / ((82.0/302.0) * 1023.0 / 1.8)) + 0.1
  print('Voltage: ' + str(val_input))

  if val_input < DETECT_VOLTAGE:
    print('-- Low Voltage --')
    print('The system may turn off due to low power input (input voltage below 4.65V), when this happens, please disconnect high power consuming peripherals or change to a qualified power supply.')

// In TinkerOS, you can find a service at "lib/systemd/system/voltage-detect.service" and similar codes "etc/init.d/voltage-detect.py".

 

Thanks.

[tkaiser]

2 hours ago, Frank Wu said:

/sys/bus/iio/devices/iio:device0/in_voltage2_raw

 

@TonyMac32 can you please check whether this commit (+ fix) is sufficient? (Still wondering why it's not just 'val2_raw / 154.31567328918322295376 + 0.1'?

[TonyMac32]

@tkaiser I was bugfixing but saw you got to it first, I assume after a reboot this should show up in armbianmonitor -m?  So far nothing.  I've verified that /sys/bus/iio/devices/iio:device0/in_voltage2_raw exists and does read values that, after the math jumble of doom, seem correct.

[tkaiser]

45 minutes ago, TonyMac32 said:

/sys/bus/iio/devices/iio:device0/in_voltage2_raw

 

Maybe needs to be used as '/sys/bus/iio/devices/iio\:device0/in_voltage2_raw' in bash scripts?

[ -f /sys/bus/iio/devices/iio\:device0/in_voltage2_raw ] && echo "Exists"

 

[tkaiser]

4 hours ago, tkaiser said:

Maybe needs to be used as '/sys/bus/iio/devices/iio\:device0/in_voltage2_raw' in bash scripts?

 

Seems so. And it seems like it's a common convention to locate those ADC sysfs nodes at this path. See 'Command to read ADC values using sysfs' in https://www.cnx-software.com/2018/08/06/aio-3399j-development-board-review-ubuntu-16-04/

[chwe]

On 8/6/2018 at 4:15 AM, Frank Wu said:

We also provide the best and official choice, the Tinker Power Supply.

the best choice lacks of some information. Thickness of cables (e.g. AWG XX). Did you check how it reacts to spikes (e.g. voltage drop when the board decides that it needs fastly 'more juice'?). Average voltage drop between low current and high current usage?

 

The 'RPi' formfactor doesn't allow sufficient powering, IMO the RK3288 is too power-hungry for a microUSB powered solution (I never had problems with mine, using an IKEA charger and a 20AWG cable but I normally don't max out consumption e.g. it runs headless, and the stuff running on it isn't high load stuff, no wifi only GbE). Maybe this gets slightly better as soon as 'The Foundation' and all those 'same formfactor boards' decide to switch to USB-C (at least the connector should be slightly better than the microUSB). 

 

On 8/6/2018 at 6:46 AM, tkaiser said:

@TonyMac32 can you please check whether this commit (+ fix) is sufficient? (Still wondering why it's not just 'val2_raw / 154.31567328918322295376 + 0.1'?

could it make sense to refractor armbianmonitor here a bit? More and more boards with 'not that dumb' PMICs are available now.. Maybe it's time to refractor it before we have a second nand-sata-install? Some sort of PMIC specific config file with paths and displayed values.. 

[Frank Wu]

2 hours ago, chwe said:

the best choice lacks of some information. Thickness of cables (e.g. AWG XX).

 

Here is this power supply's spec sheet: https://www.asus.com/Single-Board-Computer/Tinker-Power-Supply/specifications/

- DC 5.0V / 3.0A 15W

- 18AWG 150CM + Power Switch

- SCP, OVP, OCP, OTP, etc Power Protection.

- UL / CB / CE

- US DoE Level VI / EU CoC Tier 2

- LPS, etc lots of safety cert mark.

 

2 hours ago, chwe said:

Did you check how it reacts to spikes (e.g. voltage drop when the board decides that it needs fastly 'more juice'?). Average voltage drop between low current and high current usage?

 

No, currently I don't think I have kind of this data, but I can provide the result from my simple test for refer:
- low current: avg. 4.99v

- high current: avg. 4.87v (CPU 4 core, 400%.)

[chwe]

10 minutes ago, Frank Wu said:

 

No, currently I don't think I have kind of this data, but I can provide the result from my simple test for refer:
- low current: avg. 4.99v

- high current: avg. 4.87v (CPU 4 core, 400%.)

it's probably not that easy to measure.. Maybe with a oscilloscope and some advanced testing.. But that's up to the EEs here..  not my field. 

 

Did you test when a average voltage ends in instabilities under high load (surly this can't be 100% clear due to spikes). But as you may seen, we have a dedicated sub-forum for powering issues: https://forum.armbian.com/forum/31-sd-card-and-power-supply/

Might be helpful to have an 'educated guess' if *voltage pinheader* < x.xx V --> better go for a decent PSU. Still hope that those microUSB powering disappears sooner than later.. But in the meantime such hints helps us to solve such issues faster. Undervoltage can lead to all sort of 'funny' behavior some are easy, some less easy to detect. 

from: https://www.asus.com/Single-Board-Computer/Tinker-Board-S/

Quote

2. Download the TinkerOS image from the website and flash it into Tinker Board S using third-party ISO software such as Win32Disk Imager or Etcher.io (makes it easier for your users to find it).

I personally wouldn't recommend Win32Disk Imager.. Seems that the newest version is able to validate images after writing, but it's still not default... Such attempts to bring up a board often end on the same sub-forum mentioned above.. It may save you some annoying support questions (even when the majority doesn't read those recommendations).

 

[TonyMac32]

15 hours ago, tkaiser said:

Seems so. And it seems like it's a common convention to locate those ADC sysfs nodes at this path.

Hmm, something isn't pleased at either the '' in the for .. in .. do or the checks.  If I simply echo the value it works, otherwise nothing.

 

 

[tkaiser]

1 hour ago, TonyMac32 said:

Hmm, something isn't pleased at either the '' in the for .. in .. do or the checks.  If I simply echo the value it works, otherwise nothing

 

I did a quick check with some other sysfs node containing a colon and '/bin/bash -x armbianmonitor -m' execution mode. Seems like removing the single quotes should do the trick?

[TonyMac32]

I thought so, but it didn't seem to have an effect.  I'll try again later.  Removing all the conditional logic and just echoing the device value worked fine, other than formatting to 20 digits of precision (no printf in the Tinker case).

 

The removal of the single quotes works in the if, but it didn't seem to do anything in the context of the for statement, I wonder if the \ is being interpreted/passed incorrectly.

[Tido]

 

On 8/6/2018 at 3:12 PM, chwe said:

decide to switch to USB-C

USB-C connector might be a solution IF the charger and the board support power-negotiation - otherwise not !!

Therefore, I prefer DC Barrel Power Jack/Connector - it just works. No discrete electronic necessary.

 

It sounds like Frank is part of the ASUS team.

On 8/6/2018 at 4:15 AM, Frank Wu said:

We would suggest using 3A (15W)

No matter how thick your cable is, the connector is designed for 1,8A (Micro USB) and a certain amount of cycle's (plug unplug) – as a technician you should make your point towards the Designer/Marketing  how to offer at least an easy option to improve power situation for "tinkerer's".

This could be in such a way if you consider the possibilty for the customer to replace it with a DC Barrel Power Jack/Connector as it has been done here:

 

I like the Tinker board a lot, smart things like colored pins and pictorgram on the PCB – ASUS put a lot of smart ideas into that board, but missed the simple option described above.

 

Is there an engineering point of view, why such an option has not been considered ?

 

 

[Frank Wu]

8 hours ago, Tido said:

No matter how thick your cable is, the connector is designed for 1,8A (Micro USB) and a certain amount of cycle's (plug unplug) – as a technician you should make your point towards the Designer/Marketing  how to offer at least an easy option to improve power situation for "tinkerer's".y such an option has not been considered ?

 

Hi,

We've noticed this concerns at the beginning. 
As a matter of fact, believe that we are more worried about any safety design than our guests if out of the power current spec.
So that connector on the Tinker Board & Tinker Board S. We all requires our vendor need to qualified the pressure & safety stress test with 5V/3A.
They can all be relieved to used and can withstand 5v/3a.

(as I knew, the 1.8A is more like a reference standard, so also have lots connectors are required more than 1.8A.)

 

8 hours ago, Tido said:

Is there an engineering point of view, why such an option has not been considered ?

 

Since we have no intention of changing too much layout design and let it can have the highest possibility to fit RPI form-factor's 3rd party case, components, etc.

 

BTW glad to know you like our colorful design on GPIO, that is my favorite part too.

And we would be noted, also survey the DC jack design for If there are any new products in the future.

[tkaiser]

1 hour ago, Frank Wu said:

as I knew, the 1.8A is more like a reference standard

 

Are you kidding?

 

https://www.google.com/search?rls=en&amp;q=Micro-USB_1_01.pdf

 

This is the relevant specification for WHAT YOUR USERS HAVE AT HOME. The connector is rated for 1.8A max, 99.99% of cables are made for 500mA max.

 

That is the REALITY OUT THERE if you throw an electronics device on the market equipped with this crappy Micro USB connector. It encourages users to use common Micro USB gear which simply WILL FAIL with the Tinkerboard since not made for higher currents than a few hundred mA.

 

USB PD (power delivery) specs in the meantime defined NEW specifications so with both NEW receptacles and NEW cables 3A are possible with Micro USB. The USB PD specs (see here for an overview) while now defining NEW Micro USB connections (needing NEW receptacles, jacks and cables!) do of course not define any PD profile that uses 3A at 5V. Why? Since VOLTAGE DROP. Ohm's law still exists. The lower the voltage, the higher the drop.

 

The USB PD specs define Micro USB carrying 3A with NEW Micro USB connectors and NEW cables only at 12V (Profile 3) or 20V (Profile 4). And the reason is once again called VOLTAGE DROP.

 

If you put a Micro USB receptacle on your product to power your board you do two things at the same time:

• Encourage users to use the Micro USB gear that's already lying around (crappy phone chargers, average cables not suitable for more than 500mA). You introduce underpowering hassles
• Encourage users to make the wrong calculation since they think they won't need a new and special PSU if they already have a charger or an USB PSU lying around. You try to create the impression your product would be cheaper than it is (since always needing a SPECIAL Micro USB PSU which is additional costs)

By using Micro USB board makers trick their customers into believing they could re-use existing PSUs so the board appears to be a less expensive buy compared to honestly stating that while the board uses a pretty common connector it's incompatible to 99.99% of all Micro USB gear that consumers already have.

 

And the whole sh*t show started over at the RPi. If they would've started with a sane barrel plug this whole mess wouldn't exist. Since then those other board makers who copied the crappy Micro USB connector would now copy this barrel plug instead to advertise their '1:1 replacement for the RPi'.

 

Since having the same moronic discussion with the RPi fanclub some while ago: https://www.raspberrypi.org/forums/viewtopic.php?f=63&amp;t=208192&amp;start=50#p1292499 (just a quick check how many different Micro USB connectors at which amperage rating a company like RS sells. If you read through the whole thread please keep in mind that RPi forum moderators act as censors. They partially edited my posts and banned my account as a result of frequent mentions of competitor's products. But that's normal over there)

[TonyMac32]

Received an official Tinker Board supply today, will make a new topic for it, but the short version:

 

• 5.0 V 3.0 A, so no breathing room to cover voltage drop.
• 18 AWG wire size, "lamp cord" style.
• nice little switch (rocker, no illumination)
• Seems very stable under load.

Some notes on the Tinker S:

• It's reported voltage calculations are a little bit off, at 5V it reads about 0.1 volts low, and at 4.6 it is reading a bit high.  I'll have to properly probe it to verify and see if this is the case across my boards. This is most likely due to reference voltage drift on the ADC.
• The new supply configuration is set up, it introduces a large voltage drop if powering via GPIO, enough that 5.25 volts to the I/O pins reads at least 300 mV or so lower than 5.25 on the microUSB port.    After some cleaning of the terminals I got the expected results, I apologize for the false alarm

 

[tkaiser]

On 8/9/2018 at 5:39 AM, TonyMac32 said:

it introduces a large voltage drop if powering via GPIO, enough that 5.25 volts to the I/O pins reads at least 300 mV or so lower than 5.25 on the microUSB port

 

So you can neither power through GPIO any more nor use their own 'special' PSU since this thing is a 5.0V PSU and not a 5.25V as required?!

 

What do these guys think? The problem is known since day one!

 

Skip the entire review and just look at 2:19: https://www.youtube.com/watch?v=y4EIhh0VT5g#t=2m19s

 

Edit: @TonyMac32 updated his results and so still powering through GPIO header is the way to go with Tinkerboard:

 

[TonyMac32]

@tkaiser I had something messing with the GPIO pins, after a good cleaning we're back to the proper function.  I hadn't tried it on the S yet, my early test board would shut down if you tried to power it via GPIO so I was overly suspicious I think.

[LawlessPPC]

I supply extra through a usb port with a male to male cable stays at 5.12v