Core Electronics Forum

Rpi-4 usb3`s ..?...[SOLVED]

is my rpi-4 faulty referring to my other thread here


is my board faulty…??

my scattered coffe table

Hi Brian.
What is the digital multimeter measuring/monitoring?
Cheers Bob

If it’s the unit with the red led’s then that would be a power supply…the other 2 dmm`s are not connected…volts/amps/watts…

I still dont know what to think about these rpi-4 usb-3 ports.
At any rate, wether i used ubuntu or debian to set the usb boot in the eeprom …it should not matter what opp sys is on the usb media or type i would have thought…yet slower devices cvan operate on the usb 3 ports without a problem seems it is fussy about media size and or type…not shure about that either…

lsusb reveals ::

Bus 001 Device 004: ID 152d:0578 JMicron Technology Corp. / JMicron USA Technology Corp. JMS578 SATA 6Gb/s
on the usb 2 port so why wont it run on the usb 3 port

sozz i forget that usb 3 ports contain both usb2 and usb 3 but there is one protocol that usb does not have from mem…so if i just use the usb 3 ports as usb 2 only devices they work fine it is a usb3 problem…I’ll change out the cable but i dont think that is the problem…if the usb device is working ok on the usb 2 bus alone then it makes this problem so much more confusing as to where the problem lies because usb 2 devices are running fine on the usb 3 ports…hence to say this points directly to the media devices…but i have tried several and found 2 out of the 3 that i tested with did not work or were so slow I’d be better of with an sd-card, it seems that this is maybe a software issue perhaps but it`s running kernel 5.8 or 9 or something does it not have the required uass drivers built into the kernel this could be an issue i dont know…??

Hi Brian
Was just a thought. Will explain anyway. When measuring current with most digital multimeters on the milliamp ranges you should be aware of “voltage burden”. That is the higher internal meter resistance on these ranges will cause a voltage drop across the meter. This can have the effect (particularly on higher milliamp currents) of lowering the voltage actually applied to the device under test to the point where it may not work at all or behave in a different manner to normal. Using the 10 amp range (very low internal resistance) usually OK with a loss in resolution which may not matter much. Alternative is to use Hall effect or non invasive (Clamp meter) technique.
Sorry I can’t be of any help with your main problems as I know absolutely nothing about what you are doing. I just saw the multimeter there and thought if it were connected permanently you may be just falling into the trap that many others (including me) have experienced in the past and no doubt future.
Cheers Bob

1 Like

ok well i’m running 19 volts to my screen with either a buck regulator the cheap china ones i think they max at 3 amp so ok for the rpi etc and my other sbc`s having adjusted them down to 5 volts from the 19 in parallel on the same output rail… the power supply is said to be at max 30 v 10 amp… i have no doubt that what i see on those panels is not 100% correct…the supply coming from china…yes the usb 3 port issue remains a mystery unless they are indeed at fault…

i hope somebody from core can help with this problem…

the problem persists no matter what supply i am using…and yes voltage drops are big problem when using d.c and low currents…i am aware …i have analogue meters around as well i dont mind them…i use them often in my home build supply s… i`ll take a pick of one of them 4 you and the guys here…

Brian I do have one situation that may have some sort of clue.
Desktop comp. Mac Mini with 4 x USB 3 output ports.
Flight sim XPlane10 with Saitek flight yolk and throttle quadrant. CH rudder pe3dals. All USB 2.
All works fine connected directly to computer.
I have a couple of home grown add-on units to all this which I thought could put a bit of a strain on the computer USB power capability so I introduced a powered UDB hub to help and relieve the computer power supply. This is branded “Mbeat” and has 4 x USB 3 ports and 3 x USB 2 ports, all switchable.
Now, everything works fine if plugged into the USB 2 ports BUT DOES NOT WORK PROPERLY when plugged into the USB 3 ports. Yet USB 3 is OK when used directly from the computer. WHY???
Due to limited knowledge in this field I have not made any attempt to trouble shoot or fix this “problem” and am able to live with it. I just put it down to one of the quirks of this wonderful (???) digital world.
Cheers Bob

Hi Brian,

It does look like you have a few USB devices connected to your Pi. This may be affecting performance. Have you tried removing all of the USB devices.

it seems specific to the usb 3 ports…
i only have few real usb3 devices all the rest are usb 2…

a few usb sticks and a couple of hdd enclosures…these days real usb3 devices are rare it would seem to some degree…besides that i have tried with a few memory options, none seemed better than each other and my m-2 drive even worse on the usb port…which i have stated if you have read the thread from the start and also the other thread…i am not going to start another wth this issue… i have not had time to load noobs or debian onto my ram usb stick as yet 2 day …but that is my other idea,is the problem opp system responsive…i would hope not but time will tell… that only leaves poor drivers or faulty usb ports as usb 2 devices work fine in the usb 3 ports,nether the less the usb sticks i have are usb3 type sposed to be high speed… but as yet have proven not to be at all …they are sandisk ultra 64 gig usb 3 …other than that there is less support for those who dont run noob`s or debian …this helps not either.

at present i am using the 64 gig sandisk ultra it seems to work in the usb 3 port but still seems slow… the M:2 drive only works as a usb 2 device for some reason i have used it on other boards and itr works ok in the usb 3 port mode so maybe my board is faulty or there just is not support for the m:2 bridge in usb 3 mode and the device cannot decide to run in the usb 3 port correctly…t

2 things the ports or the the software …all that is left .and i cannot tell which …im in favor that there is no usb 3 support for my device in the kernel…

for you guy`s at core…you can see my lsusb data if you can be of assistance …thanks…i will post this link into the Ubuntu support page some time soon…

seems OD the usb3 sandisk has come to the party…i still think its slow as when i brought them they were supposed to be the fastest they had…i am now wondering…

i have had the m:2 to usb otg 3 bridge for some time now…it works fine with every other board i have given weather it be x86 64 or arc32 or arc 64 …but the rpi it has trouble with on the usb 3 …maybe it cannot decipher weather to run in both or a single mode …but when plugged into a usb 2 port it works fine not the fastest but it does work …faster than the sandisk in the usb 3 port…it really must be a driver issue now i think within the kernel…

furthermore to conclude …i dont think conical ever thought or rpi guys … would plug an m:2 for the op-system via the usb3 lines,…my theory…i had to push for it to be written into the kernels with other manufacturers.they were aware of it but had to wait for the kernel to be pushed uinto the system or released for thatr matter…it took time before i could use it…but on x86 64 bit it worked…

i posted on the official raspberry forum i seen a few threads concerning Ubuntu and the usb3 ports… still reading…it will take a day to be posted after admin look at it…

thought i would ad this for others that may find it help full…

sudo apt install zram-config

desktop:~$ free -t
total used free shared buff/cache available
Mem: 3829956 716448 2452408 71416 661100 2995756
Swap: 2963532 0 2963532
Total: 6793488 716448 5415940
after on my 4 gig board using the m:2 as virtual ram…

i fixed it by entering the id of the device into the cmdline.txt file it boots of usb 3 port now…seems a little faster than the usb 2 although not a lightning jump faster but faster at any rate…

so if you have a usb device that is not performing the way it should be this may well be an option to try…
code “sudo lsusb” and find the device “id” if you can see it there the number will be for E.G. “ID 1754:55as”
so insert it at the start of the line of code or after the first comment in the line at the front of the code in the cmdline.txt file as usb-storage.quirks=754:55as:u followed by the rest of the line of code and it will be loaded instead of having to rewrite a kernel or ad a patch to the kernel…etc.etc…

1 Like