Charger compatibility with the Pi 4 (Single Type C CC resistor on Pi4)

[scorpia]

I was looking at the USB C connector on the PI4 to work out what I could do with it when it arrives in the process of doing so I noticed there is only a singe cc resistor (rather than a resistor per cc channel) on the schematics. Anyone know if this is the case on the actual hardware? If it is it could cause problems with active usb cables which connect to the other cc line as well. This looks like it could make the pi and cable appear as an audio adaptor accessory as per Table 4-10 in the USB-C spec or fail to detect.

[scorpia]

Looking around the forum it appears as though this might be the cause of the issues with the pi not powering up with macbook chargers and other laptop supplies. I have written up some details into a more through posthttps://www.scorpia.co.uk/2019/06/28/pi ... resistors/ if people want more technical information.

[jamesh]

We are currently looking into why some chargers do not work. Got some on order to test.

[tvjon]

One Qualcomm QC3.0 supply I have works fine, from either its USB A or C sockets.

$ vcgencmd measure_temp
temp=54.0'C
$ vcgencmd measure_volts
volt=0.8455V


However an Anker psu also with 2 sockets, fails from its USB C output

but using an USB-A to C lead works fine from its A socket.

$ vcgencmd measure_volts
volt=0.8507V

iPad Pro psu fails, but its C to C lead which is well made, quite thin, & very flexible, so much better than the thick typical plastic-outer leads which tend to drag stuff off the bench, works on all supplies I have here. Pity I have to give it back to its owner, who is hovering here

HTH

[scorpia]

My pi has now arrived so I have been able to do some tests. My blog post above has been updated with the info as well.

I have checked and both the CC lines are wired up as in the schematic with only one CC resistor rather than two. I then checked several power supplies with and without e-marked cables (a macbook pro charger, Pixel3 charger, Office electrics desk charger). All the chargers did not power the pi when used with the e-marked cable provided with the macbook charger. when tested with a non e-marked cable all the chargers powered the pi.

Also the 5v on the connector is connected directly to the 5v rails. If the pi is powered over the GPIO then the USB-C connector is vbus hot which means it could backfeed devices when using the type C port on the pi as a USB client.

As an aside as I was testing bits I also checked the USB-C current resistors on the official charger and micro USB adaptor and both are correct.

[murratw]

My pi has now arrived so I have been able to do some tests. My blog post above has been updated with the info as well.

I have checked and both the CC lines are wired up as in the schematic with only one CC resistor rather than two. I then checked several power supplies with and without e-marked cables (a macbook pro charger, Pixel3 charger, Office electrics desk charger). All the chargers did not power the pi when used with the e-marked cable provided with the macbook charger. when tested with a non e-marked cable all the chargers powered the pi.

Also the 5v on the connector is connected directly to the 5v rails. If the pi is powered over the GPIO then the USB-C connector is vbus hot which means it could backfeed devices when using the type C port on the pi as a USB client.

As an aside as I was testing bits I also checked the USB-C current resistors on the official charger and micro USB adaptor and both are correct.

Does the official micro USB adapter have the 56k resistor? If so, how can it supply 2.5A of power? I though it would limit to 500ma. This is something i have been trying to understand.

[scorpia]

The official adaptor does indeed have the 56k resistors, this means it advertises default USB current (this includes the ability to use previous charger detection such as the battery charging spec) electrically there is nothing enforcing this on the connector side. The type C resistor is connected to the SOC I don't believe the pi is reading this.

Also the problem has been officially confirmed according to the media, cant find the original source of Ebon's comments though, if anyone has a link to that that would be useful.
https://www.techrepublic.com/article/yo ... confirmed/

In addition I made some more interesting discoveries when trying to use the pi in gadget mode. As mentioned above if the pi is powered over the gpio the USB-C port is vbus hot, this causes some devices plugged into the pi as clients to not enumerate the gadget when used with a C-C cable as they detect vbus and assume they are the client.

My phone and the 3.0 USB-C port on my laptop detect the pi and enumerate the gadget. The usb 3.1 on my laptop however doesn't. I also tried using some USB-A adaptors, I Used a USB A-C cable attached to the pi and either a USB C-A adaptor or a usb hub with a C connector to the laptop and A ports at the laptop end. With this setup the order that things are plugged in generates different results, if the pi is plugged in first it wont detect but if the laptop is plugged in first it will. This makes logical sense as if the laptop is plugged in first it goes into host mode before it sees the vbus from the pi.

[trejan]

The type C resistor is connected to the SOC I don't believe the pi is reading this.

PD_SENSE goes to one of the ADC inputs on the MxL7704 PMIC. You can read it over I2C (0x1C) but I'm not sure if the firmware actually does read it.

[bunklung]

"A smart charger with an e-marked cable will incorrectly identify the Raspberry Pi 4 as an audio adapter accessory, and refuse to provide power," says Upton.

"I expect this will be fixed in a future board revision, but for now users will need to apply one of the suggested workarounds. It's surprising this didn't show up in our (quite extensive) field testing program."

https://www.techrepublic.com/google-amp ... confirmed/

[murratw]

I make the SmartiPi Touch and am looking for a solution to split the 5v 3A USB-C input into USB-C/micro USB to power the official display. It looks like the Pi 4 does not read the 56k resistor and probably will not in the future. As the foundation offers the adapter and advertises 12.5 watts of power with it. https://www.adafruit.com/product/3187

[CedricM]

Sorry if the question seems retarded, but I'm trying to translate "e-marked cable".

Does this literally means there is an e printed on the cable?

If not what does this mean and how can one recognize an "e marked cable" ?

[Heater]

These "e" cables are not just cables. They have some kind of micro-chip in them. No idea what it does, something to do with deciding to blast your Pi with 20v and 100amps as if it were fast charging an iPhone or not.

Grief who invents this garbage? Power leads with chips in them, ridiculous.

[Burngate]

Almost as bad as putting an EEPROM on a Pi.

Or a Mondeo engine in a Volvo.

[W. H. Heydt]

These "e" cables are not just cables. They have some kind of micro-chip in them. No idea what it does, something to do with deciding to blast your Pi with 20v and 100amps as if it were fast charging an iPhone or not.

Grief who invents this garbage? Power leads with chips in them, ridiculous.

I've been nervous about USB-C ever since I found out--some years ago--that the cables are supposed to have chips in them. Obviously, there is cost factor, but we now see that a "smart" cable can be too "smart" for its own good.

[HawaiianPi]

Grief who invents this garbage? Power leads with chips in them, ridiculous.

I've been nervous about USB-C ever since I found out--some years ago--that the cables are supposed to have chips in them. Obviously, there is cost factor, but we now see that a "smart" cable can be too "smart" for its own good.

This is what happens when you try to make "the one port to rule them all" (it becomes VERY complicated).

[rpdom]

Almost as bad as putting an EEPROM on a Pi.

Or a Mondeo engine in a Volvo.

Have you done that then?

I knew someone who put a 2 litre twin cam Fiat engine in a Reliant Kitten. Not many Reliants could wheel spin from standstill in third gear and hit 125 mph in fifth

[scorpia]

e-marked cables stand for electronicaly-marked. These cables have a chip in them that can be used to query the cables capabilities over usb power delivery. A major feature that the cables advertise is their current capacity, All USB-C cables can do 3A and via the e-marking cables can state that they are able to be used at higher currents which allows for faster charging for devices such as laptops.

[HawaiianPi]

All USB-C cables can do 3A...

Sorry, but that is incorrect. There are type-C USB 2.0 cables rated for less than 3A, so no one should make that assumption.

[scorpia]

All spec compliant cables must be able to handle at least 3A (This is because there is no way to identify a non e-marked cable's current rating and 3A is available without power delivery). You might be thinking of the A-C cables and microB adaptors which need to have resistors indicating the current capacity of the attached charger is just the default USB current.

[HawaiianPi]

USB-A to C cables are still USB-C cables.

Especially in this context, since many chargers still use USB-A ports (and charger compatibility is the subject of this thread).

[chipace]

All USB-C cables can do 3A...

Sorry, but that is incorrect. There are type-C USB 2.0 cables rated for less than 3A, so no one should make that assumption.

I am sure you could make your own usb c->a cable that is rated for 1A, but the USB spec specifically calls for 3A and 5A cables, regardless of power delivery feature. All certified usb c cables can do at least 3A with a minimal drop in VBUS.
https://en.wikipedia.org/wiki/USB-C#Cables

[W. H. Heydt]

Almost as bad as putting an EEPROM on a Pi.

Or a Mondeo engine in a Volvo.

Have you done that then?

I knew someone who put a 2 litre twin cam Fiat engine in a Reliant Kitten. Not many Reliants could wheel spin from standstill in third gear and hit 125 mph in fifth

That technique used to be called a "sleeper". The favored way to do it was to remove the back seat from a VW Beetle and install a large V-8 engine.

[rpdom]

Almost as bad as putting an EEPROM on a Pi.

Or a Mondeo engine in a Volvo.

Have you done that then?

I knew someone who put a 2 litre twin cam Fiat engine in a Reliant Kitten. Not many Reliants could wheel spin from standstill in third gear and hit 125 mph in fifth

That technique used to be called a "sleeper". The favored way to do it was to remove the back seat from a VW Beetle and install a large V-8 engine.

They managed to fit the engine in the front with some custom mountings, re-routing the oil filter and making a custom exhaust system.

That engine was capable of getting the heavy Fiat up to over 100mph fairly quickly. The Reliant weighed about a quarter of what the Fiat did.

[Burngate]

Or a Mondeo engine in a Volvo.

Have you done that then?

My next-door neighbour runs a repair shop, so when my 10-year-old V70 decided to not even try to start, I asked him for help, and that's what he told me.

He also told me the fault wasn't with the engine, or the battery, or the over-complicated and far too inteligent electronics, just a broken wire from a sensor somewhere.
That would never have happened with a proper diesel - as long as there's fuel and you can turn it over, it'll run. It only needs electricity for the headlights.

[JamesPi123]

So what can I do if I want to power it from my MacBook pro charger? Use a cheap cable that is not "e-marked" or is there some funny solution I need.

By the way, what was the purpose of making a custom circuit when there was already a mandatory reference design.
https://medium.com/@leung.benson/how-to ... 470d7a5910