New Raspberry Pi model 3B+ 1.4 GHz, 330Mbit Ethernet, 802.11ac, PoE

[W. H. Heydt]

Very happy to hear USB boot is working well. Is there any provision for handling the case of more than one USB device connected at once, or does it fail if the first enumerated USB mass-storage device is not bootable?

The boot sequence documentation should answer that question. As I recall, after it checks for an SD card, it will keep checking USB devices until it finds a bootable device...or--of course--fail if there isn't one.

[W. H. Heydt]

When I ordered my first Raspberry Pi 3B it came in a paper bag (inside a cardboard box of course). My Raspberry Pi Zero W and second Raspberry 3B just came without wrap or bag (but still in a cardboard box of course). I sometimes hear people talking about anti-static bags. Maybe that's for when they ship by plane? Or they come from the other manufacturer?

The paper bag was an anti-static enclosure. The latest anti-static protection is a piece of thin cardboard that is blank other than a bar code. Over the years, I've had Pis in anti-static plastic, in anti-static paper bags, and now with the Pi3B+, just the anti-static card.

[Callous]

I ordered mine last Thursday from a website that said they had them in stock. Called them yesterday to see why they hadn't shipped yet and was told they were not in stock and would't ship until the 29th at the earliest. Than I was told that they had over 1000 units in stock for their Asian distributor and that I could get those for $10 or $11 more per unit plus a higher shipping cost. When I asked why they didn't use some of those 1000 units to fill US orders that were placed when the website said they were in stock the woman I was talking to said she had to speak to someone else about it. She came back and said to forget what she said about the other units because "That's not us, that's someone else" and just to wait but they could be delayed further. How did she know someone else had 1000 units in stock if it wasn't them?

Very shady. I think they are not shipping to people that ordered units for $35.99 while they can sell them for $45.99 each due to the high initial demand.

I usually buy them at Microcenter in Cambridge MA but figured I'd get them faster if I ordered them online rather than wait for the store to get some. I guess I was wrong.

[jamesh]

I ordered mine last Thursday from a website that said they had them in stock. Called them yesterday to see why they hadn't shipped yet and was told they were not in stock and would't ship until the 29th at the earliest. Than I was told that they had over 1000 units in stock for their Asian distributor and that I could get those for $10 or $11 more per unit plus a higher shipping cost. When I asked why they didn't use some of those 1000 units to fill US orders that were placed when the website said they were in stock the woman I was talking to said she had to speak to someone else about it. She came back and said to forget what she said about the other units because "That's not us, that's someone else" and just to wait but they could be delayed further. How did she know someone else had 1000 units in stock if it wasn't them?

Very shady. I think they are not shipping to people that ordered units for $35.99 while they can sell them for $45.99 each due to the high initial demand.

I usually buy them at Microcenter in Cambridge MA but figured I'd get them faster if I ordered them online rather than wait for the store to get some. I guess I was wrong.

You haven't said which website/distributor you were talking to, so it a bit difficult to comment. If they are misbehaving then we will give them a call. The shoudl NOT be selling above $35.

As stated elsewhere, the US supply chain lags behind a bit on a product launch since we are not allowed to import any until the FCC compliance is published, which we do on day of launch. In other regions we can fill the supply chain up front.

We are making a lot per day (well over 10k), so the supply chains should recover quickly.

[jbeale]

The boot sequence documentation should answer that question. As I recall, after it checks for an SD card, it will keep checking USB devices until it finds a bootable device...or--of course--fail if there isn't one.

Is this the documentation you're talking about? I don't see any mention of multiple USB drives. https://www.raspberrypi.org/documentati ... des/msd.md

The reason I asked was noticing this comment on a random page: https://thepi.io/how-to-boot-your-raspb ... ge-device/
"tony February 10, 2018 at 8:53 pm It works fine if I only have one USB drive attached when it boots. It doesn’t work if I have two USB drives attached."

[Callous]

I ordered mine last Thursday from a website that said they had them in stock. Called them yesterday to see why they hadn't shipped yet and was told they were not in stock and would't ship until the 29th at the earliest. Than I was told that they had over 1000 units in stock for their Asian distributor and that I could get those for $10 or $11 more per unit plus a higher shipping cost. When I asked why they didn't use some of those 1000 units to fill US orders that were placed when the website said they were in stock the woman I was talking to said she had to speak to someone else about it. She came back and said to forget what she said about the other units because "That's not us, that's someone else" and just to wait but they could be delayed further. How did she know someone else had 1000 units in stock if it wasn't them?

Very shady. I think they are not shipping to people that ordered units for $35.99 while they can sell them for $45.99 each due to the high initial demand.

I usually buy them at Microcenter in Cambridge MA but figured I'd get them faster if I ordered them online rather than wait for the store to get some. I guess I was wrong.

You haven't said which website/distributor you were talking to, so it a bit difficult to comment. If they are misbehaving then we will give them a call. The shoudl NOT be selling above $35.

As stated elsewhere, the US supply chain lags behind a bit on a product launch since we are not allowed to import any until the FCC compliance is published, which we do on day of launch. In other regions we can fill the supply chain up front.

We are making a lot per day (well over 10k), so the supply chains should recover quickly.

It was Newark/Element14. I have cancelled the order with them since I found them locally.

[W. H. Heydt]

It was Newark/Element14. I have cancelled the order with them since I found them locally.

i am astonished. I ordered from Newark mid-afternoon on launch day, the Pi3B+ was shipped the next day, and I received it on the following Saturday. The price was the official one: $35.00.

[W. H. Heydt]

The boot sequence documentation should answer that question. As I recall, after it checks for an SD card, it will keep checking USB devices until it finds a bootable device...or--of course--fail if there isn't one.

Is this the documentation you're talking about? I don't see any mention of multiple USB drives. https://www.raspberrypi.org/documentati ... des/msd.md

The reason I asked was noticing this comment on a random page: https://thepi.io/how-to-boot-your-raspb ... ge-device/
"tony February 10, 2018 at 8:53 pm It works fine if I only have one USB drive attached when it boots. It doesn’t work if I have two USB drives attached."

I will try to find where I read the boot sequence data. One thing, though...remember that there is new boot ROM code in the B0 stepping on the Pi3B+ and that should improve things. It is *supposed* to "enumerate USB devices" and test each one in turn for a bootable image.

[jamesh]

It was Newark/Element14. I have cancelled the order with them since I found them locally.

i am astonished. I ordered from Newark mid-afternoon on launch day, the Pi3B+ was shipped the next day, and I received it on the following Saturday. The price was the official one: $35.00.

Me too, sounds like someone on the phone really didn't know what they were doing. E14 would have been provided with multiple 10 of thousands of Pi's up front, so seems odd they only had 1000 left, although I can see my they might have allocations to go elsewhere since they are one of the three companies licensed to manufacture the PI worldwide.

[karrika]

I received my Pi 3B+ from Pimoroni yesterday.


To answer my own question about fitting the higher CPU in the Flirc case.
The thermal pad included is too high.


But with thermal paste instead it works just fine.

[PeterO]

I received my Pi 3B+ from Pimoroni yesterday.
But with thermal paste instead it works just fine.

Have you done a with/without paste comparison ?

I don't believe thermal paste is supposed to be used as a "gap filler". I think it is designed to improve the contact between already mating surfaces.
It might be worth checking in a few days time that the paste is staying in place and not slowly running out of the gap onto the PCB.

PeterO

[alphanumeric]

What I see in the picture is a thermal pad not past. A pad will hold the heat sink on, the ones I have used are sticky both sides. Thermal past is used "sparingly" when the heat sink is retained mechanically with a clip etc. As mentioned its not to be used as a gap filler. It is most likely conductive and bad things can happen if it gets on the circuit board. It can potentially short things out.

[karrika]

I received my Pi 3B+ from Pimoroni yesterday.
But with thermal paste instead it works just fine.

Have you done a with/without paste comparison ?

I don't believe thermal paste is supposed to be used as a "gap filler". I think it is designed to improve the contact between already mating surfaces.
It might be worth checking in a few days time that the paste is staying in place and not slowly running out of the gap onto the PCB.

PeterO

Good point. The gap is really tiny. Much less than 1mm. I could also take a knife and peel off a thin layer of a 3M thermal pad that we are using at work. At least it stays where I put it.

[drgeoff]

What I see in the picture is a thermal pad not past.

karrika's point is that the supplied thermal pad is too thick when using the case with a 3B+. That is why he used paste instead.

[alphanumeric]

What I see in the picture is a thermal pad not past.

karrika's point is that the supplied thermal pad is too thick when using the case with a 3B+. That is why he used paste instead.

Yeah, I figured that out after I posted, was distracted.

In any case, I personally wouldn't substitute past for a pad if the gap is any bigger than the thickness of a piece of paper. That's not what past is meant for IMHO.

[Imperf3kt]

What I see in the picture is a thermal pad not past. A pad will hold the heat sink on, the ones I have used are sticky both sides. Thermal past is used "sparingly" when the heat sink is retained mechanically with a clip etc. As mentioned its not to be used as a gap filler. It is most likely conductive and bad things can happen if it gets on the circuit board. It can potentially short things out.

That depends. Normally thermal paste is dielectric (insulating). Of course, that is highly dependant on what it is compounded from.
Thermal pads are usually worse than thermal paste at transferring heat. They're really only just more convenient as they don't require anything to hold them on.

E: Wikipedia's article is actually better than I am at explaining it.
https://en.wikipedia.org/wiki/Thermal_grease

[rv112]

I got my RPi 3 B+ now running with LibreELEC on it and without any heatsinks. After 10 minutes in idle I get:

Code: Select all

LibreELEC:~ # vcgencmd measure_volts core
volt=1.3500V
LibreELEC:~ # vcgencmd measure_temp
temp=62.3'C

Do you think this can get a problem? What would be the best heatsink option? I have 2 small heatsinks but there is no space for the CPU on the backside in my case.

[0xFF]

Is the new model hackable for the external wifi antena?

[jamesh]

I got my RPi 3 B+ now running with LibreELEC on it and without any heatsinks. After 10 minutes in idle I get:

Code: Select all

LibreELEC:~ # vcgencmd measure_volts core
volt=1.3500V
LibreELEC:~ # vcgencmd measure_temp
temp=62.3'C

Do you think this can get a problem? What would be the best heatsink option? I have 2 small heatsinks but there is no space for the CPU on the backside in my case.

That temperature is fine, well below any sort of problem level (80+).

[jamesh]

Is the new model hackable for the external wifi antena?

Not that I am aware of, and we cannot condone doing it as it contravenes the FCC and EU compliance.

[bensimmo]

I've just grabbed mine to have a look.

.. if you look near the UK text there is a square pad. Spending some time (I'm in a dark room at the moment) there is a flint on the antenna line to the cavity/metal cover right next to it.
My guess is you can bridge that, cut the line (or remove the first white antenna component).
You fit a uFL to that square and remove some of the circuit board green cover and solder the larger pads to hold/earth it in place.

Though I've not looked, there will probably already be somebody with a detailed mod to do it. As is usually the way.

If not, if you do it and it works, write a blog on it.

Edit grabbed a photo, handy these phones.

PhotoPictureResizer_180323_175034762_crop_439x460.jpg (166.87 KiB) Viewed 10296 times

[hippy]

Is the new model hackable for the external wifi antena?

Looks like it may be -

https://fccid.io/png.php?id=3777895&page=4

[mahjongg]

Is the new model hackable for the external wifi antena?

Looks like it may be -

https://fccid.io/png.php?id=3777895&page=4

That's not an external antenna, its a connection to a spectrum analyzer.
So the "flint on the antenna line to the cavity/metal cover right next to it." is there NOT for an external WiFi antenna, but for a spectrum analyser.
Again, the RPF does NOT CONDONE USING AN EXTERNAL ANTENNA!

[bensimmo]

Doesn't stop anyone modding it and trying.
Just like the ZeroW and Pi3, they have the same testing use. As per Raspberry Pi Trading information.

As per many of the RaspberryPi Foundations HackSpace/MagPi would encourage you to do.

[hippy]

Is the new model hackable for the external wifi antena?

Looks like it may be -

https://fccid.io/png.php?id=3777895&page=4

That's not an external antenna, its a connection to a spectrum analyzer.
So the "flint on the antenna line to the cavity/metal cover right next to it." is there NOT for an external WiFi antenna, but for a spectrum analyser.

Isn't it what they call a "uFL to SMA pigtail" or similar, which the spectrum analyser then connects to ?

The accompanying FCC test report may describe it in detail. I didn't check.

Whatever the connection is there for - and it almost certainly was included only so testing can be done exactly as shown in the photo rather than for connecting an actual external antenna - it doesn't mean it cannot be used for something else beyond its original intent.

I'm not saying it can, only said that it may.

Again, the RPF does NOT CONDONE USING AN EXTERNAL ANTENNA!

Yes I know that, I have heard that, I accept that. No need to shout.

I am not encouraging anyone to do it, just saying it may be possible to do it, just answering the question asked.