Specific question: Will a 65W PD3.0 charger be able to work with a FW16 with no dGPU under light/moderate workloads? Will it charge while on or not on? Or will it not work at all?
General question: Is there any official guidance from framework on what specific charger requirements are or are supported? Will any PD3.0 charger charge battery? Is PD3.1 required? I realize actual power draw varies based on options and workloads, but some general guidance would be really helpful.
Welcome to the forum.
If you don’t have a FWL16 yet, I’d strongly recommend getting the charger with it. There have been some issues with other chargers. Other chargers can work, lower wattage can work, but it’s not assured.
Framework recommends at least 100W without dGPU. With it, 180W is the minimum. It also the maximum as nothing more is available on the market.
I (without dGPU) am using a Baseus 60W charger with no issues, the laptop limits itself to 60W - so I’m losing about 20% performance. Whatever is leftover when not under full load gets put into the battery.
I also have a 30W charger. With that it doesn’t charge when on and might pull from the battery under load. It does charge when off.
Don’t have an FW16, however AFAIK when using FW13 plugged in an underpowered charger it just draws constant power almost all the time and the higher the load the lower the changing rate. The transition to and from discharging from and to charging is seamless
I have done some tests with my 65W charger and I wouldn’t recommend it. The performance will be worse while charging as the CPU then gets limited to 30W instead of 45W which it would get with a higher wattage charger or on battery under max performance. So either get a 100W or the Framework charger as the usb 4 dock I’m using charges at 100W and the CPU reaches 45W.
Blockquote Framework recommends at least 100W without dGPU. With it, 180W is the minimum.
Is this on the website somewhere? Can you point to any offficial statement? Also is there a specific USB PD spec (such as PD3.1) that is required? Thanks!
It’s in the FAQ of the configuration step:
The FW16 does not charge with 5V, it also doesn’t do Quick Charge(QC).
I can confirm that the Razer Core X works for charging the FW 16.
That bit is potentially a bug like it is on the 13 and is supposed to work but currently kinda borked. The hardware is capable and can be tricked into doing it.
Has any (non qualcom arm) laptop ever worked with quick charge?
I kinda hoped the FW16 would come out with a less borked PD implementation, and so far it has worked with anything I own that counts as an Laptop power supply. Including my Lenovo dock and my Dell monitor that provides PD. Unfortunalty they are all 60W, but I got the FW charger.
Probably not, but I thought it would be a good place to state it.
Pretty sure the 16 shipped with the fixed ec firmware that fixed the power sequencing thing, the charging of 5v is still not working on the 13 or the 16 without tricking the laptop into it.
In my testing all PD chargers have worked on my FW16 (I don’t have a dGPU) including 30w, 45w, 65w, 100w, and 180w chargers.
With the 65w charger (from Lenovo) I have observed that the system seems to automatically limit its performance and power draw in order to avoid draining the battery. This results in an ~20% drop to multi core performance.
With the 30w and 45w chargers (both are from Anker) if I use the laptop while charging the battery sometimes drains (although if I’m just browsing the web they are able to keep the charge level mostly stable).
With the 100w charger (I think it’s from UGREEN) I get full performance without draining the battery (but if I’m using it while charging it doesn’t charge at full speed).
With the 180w charger (from Framework) I get full performance and the laptop charges at maximum charging speed simultaneously.
I have a PinePower power supply, which has a 65W USB-C port that implements 20V3.25A over PD.
The Framework seems to be limiting itself to 20V2.5A no matter under what load or state. This either seems like a firmware issue or issue with the PinePower not advertising its PD modes correctly. This is pretty annoying, since with only 45W it can sometimes use up battery under heavy load.
Did anyone have similar experiences?
Yeah it’s kinda annoying having to unplug the charger to get full multicore performance (PPT of 40W for a bit then 30W sustained) instead of the 20W max PPT I get with a 65W charger.
For references, this is can be found in the UMA_AC_DC_PMF tables in the pmf.c files if somebody wants to tune this in the EC FW: 7000 Series, AI 300
Notably, firmware treats a 65W charger the same as even a 35W charger, but even then, I don’t understand if there’s an limitations in the charging circuit that would prevent it to at least get the performance we get on battery, at least on “best performance” profile - since that doesn’t have the expectation of still charging the battery as you may have with a “best efficiency” on a 45W charger.
struct pmf_info UMA_AC_DC_PMF[] = {
...............
{80, EC_AC_BEST_PERFORMANCE, 43, {227, 65, 54, 45, 0}},
{80, EC_AC_BALANCED, 44, {227, 65, 48, 40, 0}},
{80, EC_AC_BEST_EFFICIENCY, 45, {221, 63, 42, 35, 0}},
{1, EC_AC_BEST_PERFORMANCE, 46, {202, 58, 38, 32, 0}}, // This is gimped performance on <80W chargers
{1, EC_AC_BALANCED, 47, {189, 54, 36, 30, 0}},
{1, EC_AC_BEST_EFFICIENCY, 47, {176, 50, 34, 28, 0}},
};
struct pmf_info UMA_DC_ONLY_PMF[] = {
{0, EC_DC_BEST_PERFORMANCE, 62, {85, 60, 40, 30, 0}}, // This is full performance on battery
{0, EC_DC_BALANCED, 63, {85, 53, 35, 30, 0}},
{0, EC_DC_BEST_EFFICIENCY, 64, {85, 45, 30, 30, 0}},
{0, EC_DC_BATTERY_SAVER, 65, {85, 45, 30, 30, 0}},
};
So I still think this is a bug (initially flagged here), but I don’t have the time to get into this and fix it, sadly.
EDIT: Took the wrong PMF, the difference is not as stark anymore. Not sure if this PMF alone can explain the huge gap in PPT I see. When I initially flagged it (see above), it was in the thermal limit section, so maybe the true culprit is found elsewhere after all.