User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Model A, A+ Issues. (Solved)

Thu Nov 13, 2014 3:01 pm

I've posted about this before but didn't get any meaningful replies. There are differences between what works in a B/B+ USB and what works in an A. Just spent most of the day, trying to determine why a well trusted RTL8188CUS wifi adaptor works intermittently in my new A+. It works perfectly in any B or B+ I plug it into with all the other constants remaining the same. Same SD card, with the same code, swapped in and out of an adaptor to try in old school A/Bs 3 different power supplies all of which power B/B+ perfectly, one of which is the 'holy' genuine Swag store 2A extravaganza :
Image

All of them give the right side of 5 volts at the header pins on the A's . This being Europe, all the power supplies have fixed leads.

3 different wifi adaptors which all happen to be RTL8188CUS chipsets, all work perfectly in B's. In the A's they are problematic, and each of the 3 adaptors varies in how problematic. 2 of them work all the time in my 'A' the third is intermittent, sometimes it works and sometimes it doesn't. SSHing to it will not work one minute, and gives the old 'no route to host' message and 2 minutes later it will connect and I haven't touched anything. When its connected it doesn't respond cleanly I type something and there is often several seconds between typing and it appearing on the screen. a minute later it works normally.

On the new A+ all three of them are intermittent with the worst on the A refusing almost all of the time, with the other 2 working sometimes. Identical symptoms to the A. The A with one of the 'good' adaptors will work long term. several weeks I've had it up.

Jumped all the hoops, latest Raspbian, numerous incarnations of /etc/network/interfaces and other configuration options, none of which makes any difference.

I've had RT5370 adaptors in the past but it looks like I sold them to customers (they seem to work better with a certain Redmond OS), I've ordered another to see if it works with the As.

This is my usual stuff that works fine on all B's and the A+ sometimes :

Code: Select all

sv@STSR-Desktop:~$ ssh root@192.168.1.253
Linux raspa 3.12.28+ #709 PREEMPT Mon Sep 8 15:28:00 BST 2014 armv6l

The programs included with the Debian GNU/Linux system are free software;
the exact distribution terms for each program are described in the
individual files in /usr/share/doc/*/copyright.

Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
permitted by applicable law.
Last login: Thu Nov 13 12:35:51 2014 from 192.168.1.102
root@raspa:~# cat /etc/network/interfaces
auto lo
iface lo inet loopback
auto wlan0
iface wlan0 inet static
address 192.168.1.253
broadcast 192.168.1.255
gateway 192.168.1.1
wpa-ssid "TP-LINK_xxxxx"
wpa-psk "whatever"
wireless power off

root@raspa:~# lsusb
Bus 001 Device 002: ID 0bda:8176 Realtek Semiconductor Corp. RTL8188CUS 802.11n WLAN Adapter
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
root@raspa:~# 
Before anyone asks, I have tried rpi-update, I avoid it usually. It had no effect, so I put it back to stock. I'm a great believer in purging ifplugd since it causes more problems than it fixes. As should be clear from lsusb, there are no hubs in the eqaution with the As.

Maybe the developers can shed some light. All suggestions gratefully received.

I've been around long enough with enough posts to show I'm no noob. Been around Linux since the 90's, I like to think I have a clue and I earn a living mending computers.
Last edited by pluggy on Fri Nov 14, 2014 11:00 pm, edited 1 time in total.
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

simplesi
Posts: 2327
Joined: Fri Feb 24, 2012 6:19 pm
Location: Euxton, Lancashire, UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 3:39 pm

Have you tried switching it off and on again........
Seeking help with Scratch and I/O stuff for Primary age children
http://cymplecy.wordpress.com/ @cymplecy on twitter

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 3:44 pm

simplesi wrote:Have you tried switching it off and on again........
Image
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

Ravenous
Posts: 1956
Joined: Fri Feb 24, 2012 1:01 pm
Location: UK

Re: Model A, A+ Issues.

Thu Nov 13, 2014 3:48 pm

Weird. Possibly some electrical issue on some of your A boards like inconsistent or out of tolerance capacitors? (I mean the "biggest" one on the power input.) Just a completely wild guess.

You're talking about As aren't you, not A+? (I ask because I think the A is exactly the same board as the B, i.e. the same traces and everything. I can't see how an electrical difference occurs with the same board, unless the lower current draw of the A is a partial cause.)

BMS Doug
Posts: 3823
Joined: Thu Mar 27, 2014 2:42 pm
Location: London, UK

Re: Model A, A+ Issues.

Thu Nov 13, 2014 3:54 pm

Ravenous wrote:Weird. Possibly some electrical issue on some of your A boards like inconsistent or out of tolerance capacitors? (I mean the "biggest" one on the power input.) Just a completely wild guess.

You're talking about As aren't you, not A+? (I ask because I think the A is exactly the same board as the B, i.e. the same traces and everything. I can't see how an electrical difference occurs with the same board, unless the lower current draw of the A is a partial cause.)
there is very specifically a difference between the A and B where USB is concerned.
Doug.
Building Management Systems Engineer.

Ravenous
Posts: 1956
Joined: Fri Feb 24, 2012 1:01 pm
Location: UK

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:00 pm

Duh - OK to summarise (my little brain isn't well today) the B has the extra USB chip and the A and A+ are both having the problem (some or all of the time) but one of the A boards is fine. Is that right?

Ignore my comments about the power regulation anyway. Got to be something about the SoC's USB output directly, hasn't it. (In combination with the three wi-fi adapters mentioned.)

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:01 pm

I have all the versions and models of the Pi , except the CM. (I don't have the revision 1 with the zero ohm resistors either if we're getting very picky).

I wouldn't think that too much current capacity would be an issue. USB has always been problematic, especially in the early days, I suspect the much lower number of 'A's around means it hasn't had quite the workout and checking that the 'B' has had. Not having a fixed hub in the loop is bound to have consequences. Why the A should be different to the A+ is anybodies guess.

The 2A power supply would probably feed 10 A+ with cheap mini Wifi adaptors at the same time......
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

mikerr
Posts: 2826
Joined: Thu Jan 12, 2012 12:46 pm
Location: UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:06 pm

FWIW I'm using a Ralink 5370 in the A+ without issue:

Code: Select all

lsusb
Bus 001 Device 002: ID 148f:5370 Ralink Technology, Corp. RT5370 Wireless Adapter
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
(I'm actually running it at <5v it because I don't want the red LED )

the small size of the A+ make the usb dongle look bigger, so I might swap to a nano wifi.
Image
Last edited by mikerr on Thu Nov 13, 2014 4:12 pm, edited 2 times in total.
Android app - Raspi Card Imager - download and image SD cards - No PC required !

Ravenous
Posts: 1956
Joined: Fri Feb 24, 2012 1:01 pm
Location: UK

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:08 pm

pluggy wrote:Why the A should be different to the A+ is anybodies guess.
I read your original post as meaning the A and the A+ are almost the same - don't work - just the exception that ONE of the A boards is fine. Correct me if I'm wrong though.

I was thinking power supply regulation, because I know dodgy PSU's don't always regulate well at particularly low currents. The capacitor might influence the stability of regulation. But that's probably not it, as Doug said the obvious difference is the B's USB chip itself.

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:13 pm

mikerr wrote:FWIW I'm using a Ralink 5370 in the A+ without issue:

Code: Select all

lsusb
Bus 001 Device 002: ID 148f:5370 Ralink Technology, Corp. RT5370 Wireless Adapter
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
(I'm actually running it at <5v it because I don't want the red LED )
If I can get one type to work reliably it will do. Should have hung onto the last 5370, but I've a family to feed.....

I seem to remember the last one I sold was actually a 8188CUS but the customer was having problems with it and I swapped if for the 5370.

There's another one in the post somewhere.
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:22 pm

Ravenous wrote:
pluggy wrote:Why the A should be different to the A+ is anybodies guess.
I read your original post as meaning the A and the A+ are almost the same - don't work - just the exception that ONE of the A boards is fine. Correct me if I'm wrong though.

I was thinking power supply regulation, because I know dodgy PSU's don't always regulate well at particularly low currents. The capacitor might influence the stability of regulation. But that's probably not it, as Doug said the obvious difference is the B's USB chip itself.
I have one A and one A+ and half a dozen various model 'B' s. The A works OK with 2 of the adaptors, the A+ has problems with all 3. None of the Bs have any problems with any of the adaptors.

Looking at my '+' boards, capacitors don't seem to be in abundance. The big 'thumb rest' certainly isn't there.

The official Swag store 2A supply HAS to be beyond reproach, hasn't it ?
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

Ravenous
Posts: 1956
Joined: Fri Feb 24, 2012 1:01 pm
Location: UK

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:29 pm

You're right - forget what I said about the power supply. I was just looking for possible reasons why the Bs are all fine but BOTH of your A-types are at least partly dodgy.

Needs an opinion from one of the foundation's developers really, as you tried to tell us in the first place :(

User avatar
mikronauts
Posts: 2803
Joined: Sat Jan 05, 2013 7:28 pm
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:31 pm

I'd measure the 5V at the GPIO connector, and see if it is being pulled too low by the WiFi adapter.

Next, I'd measure the 3.3V - perhaps there is something wonky with the 3v3 regulation on your board?
http://Mikronauts.com - home of EZasPi, RoboPi, Pi Rtc Dio and Pi Jumper @Mikronauts on Twitter
Advanced Robotics, I/O expansion and prototyping boards for the Raspberry Pi

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:44 pm

mikronauts wrote:I'd measure the 5V at the GPIO connector, and see if it is being pulled too low by the WiFi adapter.

Next, I'd measure the 3.3V - perhaps there is something wonky with the 3v3 regulation on your board?
Using the 'holy' Swag Store PSU, The A board reads 5.10 volts on the header pins, the A+ reads 5.12 volts, both boards read 3.30 volts dead across the 3.3 volt pins.

Don't make me dig the 'scope out.......
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

BMS Doug
Posts: 3823
Joined: Thu Mar 27, 2014 2:42 pm
Location: London, UK

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:46 pm

pluggy wrote: Why the A should be different to the A+ is anybodies guess.

I'd guess that the clock speed signal for the B+ has changed the USB setup significantly, so the A+ will vary from the A in a corresponding fashion. I'm pretty sure this is all going to be an issue that will require a kernel re-write.
Doug.
Building Management Systems Engineer.

PaddyC
Posts: 27
Joined: Thu Aug 28, 2014 5:55 pm
Location: Maine, USA

Re: Model A, A+ Issues.

Thu Nov 13, 2014 4:58 pm

Just a guess, could this be related?
viewtopic.php?f=63&t=38569

I have 4 of the mentioned wifi adapters, and 3 model As. All 3 Pi's were affected. The only way I could run wifi was via powered hub. The issue I had was 100% the broken ground pin on USB. It would work some/most of the time directly attached but solid as a rock via hub.

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 5:12 pm

Its useless for my purposes if it needs a hub. It does behave differently if I plug a hub into the equation. If I wanted more than 1 USB I'd use a B+ and the issue vaporises into the ether.

Headless server running GPIO connected peripherals, has to be cheap and compact.

Doug's explanation sounds like a very likely cause. If the RT5370 works without issue as Mikeerr implies when it arrives I'll live with it.
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

jdb
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 2415
Joined: Thu Jul 11, 2013 2:37 pm

Re: Model A, A+ Issues.

Thu Nov 13, 2014 5:56 pm

Please post a dmesg log from boot that encompasses the problem behaviour of the wifi dongle(s).
Rockets are loud.
https://astro-pi.org

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Thu Nov 13, 2014 5:59 pm

jdb wrote:Please post a dmesg log from boot that encompasses the problem behaviour of the wifi dongle(s).
OK, gimme a while..... I'll need to get it in the half working mode so the Pi knows about it, I suspect its going to be none the wiser in the 'no route to host' moments.
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Fri Nov 14, 2014 1:19 pm

Bahhh, Still on this. When you want it to half work, its either entirely dead or working perfectly......

Either way you don't get any clues in dmesg.
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

mikerr
Posts: 2826
Joined: Thu Jan 12, 2012 12:46 pm
Location: UK
Contact: Website

Re: Model A, A+ Issues.

Fri Nov 14, 2014 1:52 pm

All Model A or A+ I've used seem to have a "tight" USB socket,
requiring a fair amount of force to insert/remove a device

I've seen people on here say likewise too
- it could possibly cause a break in a weak/poorly soldered USB dongle

Notably the Model B/B+s sockets don't have that problem.
Android app - Raspi Card Imager - download and image SD cards - No PC required !

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Fri Nov 14, 2014 2:57 pm

I've pulled the top off a Wifi adaptor unplugging it from an A+. It is tight, but not caused any electrical issues.

The Bs have double sockets and there's invariably more room in them.
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

jdb
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 2415
Joined: Thu Jul 11, 2013 2:37 pm

Re: Model A, A+ Issues.

Fri Nov 14, 2014 4:02 pm

Do you have a (short) Type A plug to Type A USB socket lead (i.e. extension cable)?

If so then try using that as an interposer between the A+ and the adapter.
Rockets are loud.
https://astro-pi.org

User avatar
Mettauk
Posts: 238
Joined: Mon Dec 10, 2012 12:40 pm
Location: Zarg

Re: Model A, A+ Issues.

Fri Nov 14, 2014 4:55 pm

OK clear issue that does not seem to be power related.

Pi camera with servo pan & tilt head. Latest noobs Raspain with all updates to 14 Nov 14 installed on an A+ and on an A

Pan and Tilt will not work via a web interface with or without seperate power supply for servos or with OpenElectrons board on the A+ Pi. It will work via python though. So something I dont understand is going on. :cry:

On boot up on the A+ at Login I get the following error message;

Code: Select all

Login: Neutral the pan and tilt servos
Traceback (most recent call last):
File "/home/pi/pi-pan/pipan_pipe.py", line 9, in <module>
p_ind = pilight.PILIGHT()
File "/home/pi/pi-pan/pipan_pipe.py", line 61, __init__
OpenElectrons_i2c.__init__(self, pilight_address >> 1)
File "/home/pi/pi-pan/OpenElectrons_i2c.py", line 54, in __init__
self.bus = smbus.SMBus(b)
[0Error: [Errno 2] No such file or directory
done
Thios error message does not happen on the A or B boards I have.
Any sugestions?
As humans we have been the same for a very very long time, technology changes how we do... not who we are as people.

User avatar
pluggy
Posts: 3635
Joined: Thu May 31, 2012 3:52 pm
Location: Barnoldswick, Lancashire,UK
Contact: Website

Re: Model A, A+ Issues.

Fri Nov 14, 2014 7:11 pm

jdb wrote:Do you have a (short) Type A plug to Type A USB socket lead (i.e. extension cable)?

If so then try using that as an interposer between the A+ and the adapter.
Somewhere, my daughter decided to 'tidy up' my desk and I can't find anything.

It seems if I leave the Pi running for a while even though I can't get into it, it goes through the half working phase, and after a couple ,of hours it seems to come right and continue working. Last night after failing to get to talk to it, I left it and watched the telly for 2 hours, when I came back it logged in perfectly. I left it overnight and tried it this morning, it was still OK. Left it all today while I was out on a job and tried it again just now. By this time the Pi had been up for 22 hours, and it was still working perfectly. So I rebooted the Pi and bang, its immediately back to "No route to host". I'll have a dig and see if I can find the USB extension lead. The RT5370 adapter showed up today so I've that to try as well.

Code: Select all

sv@STSR-Desktop:~$ ssh root@192.168.1.253
Linux raspa 3.12.28+ #709 PREEMPT Mon Sep 8 15:28:00 BST 2014 armv6l

The programs included with the Debian GNU/Linux system are free software;
the exact distribution terms for each program are described in the
individual files in /usr/share/doc/*/copyright.

Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
permitted by applicable law.
Last login: Thu Nov 13 21:55:47 2014 from 192.168.1.102
root@raspa:~# uptime
 18:27:33 up 22:10,  1 user,  load average: 0.00, 0.03, 0.05
root@raspa:~# uptime
 18:55:37 up 22:38,  1 user,  load average: 0.04, 0.03, 0.05
root@raspa:~# reboot

Broadcast message from root@raspa (pts/0) (Fri Nov 14 18:56:11 2014):

The system is going down for reboot NOW!
root@raspa:~# Connection to 192.168.1.253 closed by remote host.
Connection to 192.168.1.253 closed.
sv@STSR-Desktop:~$ ssh root@192.168.1.253
ssh: connect to host 192.168.1.253 port 22: No route to host
sv@STSR-Desktop:~$ ^C
Don't judge Linux by the Pi.......
I must not tread on too many sacred cows......

Return to “General discussion”