I have here two "Arduino Nano v3" boards by Gravitech US and neither of them are being recognized by my computer over USB. When I run lsusb nothing shows up. I can, however, program them if I use the ICSP connector or by connecting a separate UART board to Tx/Rx/Reset.
The protection diode on both of these was failed. Since I could program the 328P, I replaced them and the power LED now lights up. I am a little baffled by the regulator. I had assumed to be an LM1117 as per the official Arduino circuit diagram, but the pinout seems different. The LM1117 has its output on the centre pin (2) and top tab while this part appears to have the output on pin 3 and GND on the centre pin. The part is marked 32RG NOOA', the other is slightly different being marked 83AK NOOA`.
Is anyone familiar with this board and might know what regulator has been used here?
Second question is, does anyone have a circuit diagram of this board?
I was not able to find a circuit diagram for this Gravitech version of the board.
The fact that it is not showing up as a USB device suggests that the FTDI chip is probably failed on both boards which is unfortunate. I do see VCC in on pin 20 and 3.3V on the 3V3OUT pin 17 and the Tx/Rx LEDs do flash several times (6?) when the board is plugged in. I do have a spare salvaged FT232RL lying around somewhere so I might have a go at swapping it out on one of these boards, but before I do, was just wondering if anyone might have other ideas as to what might be going on here?
BTW, the protection diodes having failed might mean that neither of these boards can be trusted anyway so there is that to take into account.
Verify the USB cable is for DATA.
Try the data/Nano on a Windows computer.
I have a handful of Nano V3 with USB-C... they all work. US$3 each from Amazon.
Fair enough, I hadn't stated that specifically, however, I can confirm now that the cable is definitely for data and works just fine with the separate orange FTDI breakout board board.
The only ones I have had fail were ones with fake FTDI chips. Never really had a problem with them otherwise. However these are genuine but used boards that came in a job lot and I am beginning to question the reliability of the source.
Dmesg reports as follows:
[ +10.093209] usb 3-3: new full-speed USB device number 12 using xhci_hcd
[ +0.120095] usb 3-3: device descriptor read/64, error -71
[ +0.224135] usb 3-3: device descriptor read/64, error -71
[ +0.224068] usb 3-3: new full-speed USB device number 13 using xhci_hcd
[ +0.120136] usb 3-3: device descriptor read/64, error -71
[ +0.224061] usb 3-3: device descriptor read/64, error -71
[ +0.104167] usb usb3-port3: attempt power cycle
[ +0.392095] usb 3-3: new full-speed USB device number 14 using xhci_hcd
[ +0.000118] usb 3-3: Device not responding to setup address.
[ +0.204076] usb 3-3: Device not responding to setup address.
[ +0.203992] usb 3-3: device not accepting address 14, error -71
[ +0.000051] usb 3-3: WARN: invalid context state for evaluate context command.
[ +0.120017] usb 3-3: new full-speed USB device number 15 using xhci_hcd
[ +0.000112] usb 3-3: Device not responding to setup address.
[ +0.204132] usb 3-3: Device not responding to setup address.
[ +0.204025] usb 3-3: device not accepting address 15, error -71
[ +0.000054] usb 3-3: WARN: invalid context state for evaluate context command.
[ +0.000067] usb usb3-port3: unable to enumerate USB device
So as it turns out it is detecting that something was plugged in, but is unable to read it. I did trey on Windows and I get:
Which is kind of the same thing, just not in so many words.
Were these actually manufactured by Gravitech for Arduino? I just found this link on the official Arduino site that has a picture of what looks like the same board:
BTW, I did wonder whether the FTDI chips on these Nanos might have been "bricked" by FTDI at some point. Comparing the board with the photo on the Arduino website shows some minor differences such as a slightly heavier font, the 'G' logo is different, and just above the USB connector my board says "USA 2013" rather than "USA 2009" and the '2013' is not level with the 'USA'. The colour is also slightly darker, although its difficult to be sure with photos. This is enough to at least suspect that it might not be a genuine board.
However, these FTDI chips do not respond with a PID of "0000" like the "bricked" ones are described as doing. As per messages in the previous post, the OS just shows an error. The FTDI_Prog.exe program for Windows does not find them, so I am not able to re-program them with the correct PID even if they had been bricked.
Either way it seems the chips might be corrupt of faulty in some way.
To me, it looked somewhat like even (some?) genuine Gravitech Nanos might have ended up with fake FTDI chips...
(realize that in the early days, the FTDI chips were essentially the most expensive and hardest to get component on an Arduino.)
If you are confident with your skills, you can certainly replace the FTDI chip. A USB device not able to return device descriptor is dead. I remember there was a period of time around 2016(?) when a nano on aliexpress cost less than an ATMEGA328 from digikey. There are certainly questionable parts on the market. If the project is important, you may want to get a genuine nano.
I guess that is correct and these boards have suffered some serious abuse and the USB problems are only a symptom of that. I've destroyed Nanos by probing for voltages with a multimeter while failing to notice that the leads were set up for the 10 amp range. The tiny Scottky protection diode emits a tell tale puff of smoke and even immediately removing the leads is too late.
No project at the moment. Just trying to determine whether these boards can be salvaged at least for general messing about with. If these were definitely clones from China then they might have gone straight in the bin. However, I figured that since these appeared to be genuine, although perhaps 3rd party boards, they might be worth a little effort. I suspect you are correct about the UART chips though.
One of them had a telltale scorch mark, but the other seemed to have no external evidence of damage, yet both had failed.
I can not guarantee that it's what you're looking for.
Gravitech (as far as I know) designed the Nano; the schematic for version 3.2 specifically states that 3.2 is based on a design by Gravitech. I do not know if they produced Nanos; maybe the silkscreen just has their name on it?
Thank you. This does confirm that this version of the Nano had a different regulator. The circuit shows a UA78M05 and the data-sheet confirms that this does have a different pinout to the AM/LM1117.
The boards can be programmed using the ICSP header. Since I also ended up with five ShiftBrite modules and have not seen these before, I was curious to see whether they work, so used one of these Nanos to drive them. I couldn't find a ShiftBrite library in the Arduino IDE, but I did find a sketch online that bit-banged the IO pins directly. It turns out that both the sketch and the LEDs do work. It looks like they have the three RGB colours, but no white. Would these have been an early version of the WS2812 LEDs that we have today?
(The Teensy board on the left is part of a separate project and power if from an external supply.)
Yeah, the ShiftBrite modules (from MaceTech) were an early attempt to ease the connection of "many" RGB LEDs to microcontrollers. I still have some too; bought them at a Maker Faire and haven't used them. There are some other similar historical oddities, notably the "voxels" used in Mark Lotter's "Cubatron" at Burning Man, back in the pre-Arduino days (2004!)
Presumably, some of the early super-large LED TVs (at ball parks and etc) had some sort of pre-ws2812 drivers as well.
Well, I made an attempt to repair the first of the Gravitech Nanos, but had a slight mishap during removal of the existing FTDI chip, in that four of the pads came off. On further examination it turned out that these pads were not actually connected anywhere so after carefully cleaning up the remaining pads, I decided to go ahead and solder on a new FTDI chip. Thankfully after burning the bootloader, I was able to successfully upload the Blink sketch. One down, one more to go.
The repair on the second one went without a hitch. After replacing the chip, the bootloader was burned and that one also then successfully accepted the Blink sketch over USB.
The reason for having to burn the bootloader was because I had previously tested programming the boards using their ICSP header, which overwrites the bootloader.
My effort seems to have paid off and these two Gravitech Arduino Nano boards have now been repaired. Had they been clone boards, they would simply not have been worth the effort and would have just gone in the bin. As it is, since these were still programmable via the ICSP header and original, I figured that it was just maybe worth a shot, if only for the soldering practice. However, I would not recommend it unless you have the appropriate tools and at least some experience of soldering SMD parts.
