Why not USB host option in the Xmega range.

Go To Last Post
16 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Given the rich set of peripherals found in the great Xmega range, why o why do I have to drop back to the good but limited AT90SUB1287 to get host USB functionality. If there are any decision makers from Atmel reading this - believe it or not,  there is a real need to interface to USB slave devices from a small low power host. The Xmega range is really very versatile but sorely misses this feature.  So please, please, consider putting in this feature as the use of  larger SAM/ARM chips is unnecessary bloat.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jalbinson wrote:

.... as the use of  larger SAM/ARM chips is unnecessary bloat.

Many may agree, but not Marketing Departments - if you want to avoid ARM, look at FTDI hosting options ?

Or, if you want simpler/smaller, Digikey shows Freescale, Microchip, and Renesas all with sub $2/1k USB OTG offerings.

 

Interesting to note, that all of the lowest price offerings are 32b

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I was surprised that the E5 range didn't have USB at all. I wonder if it is a marketing thing or if there is some technical reason for not including it. Host/OTG does add some complexity, and of course  they would need to port their software stack, but that doesn't explain the lack of a USB device peripheral on E5. Cost perhaps, but the U versions of other parts don't seem to be any more expensive.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
... but that doesn't explain the lack of a USB device peripheral on E5.
The XMEGA E ADC has some improvements (source impedance range, 16bit mode); therefore, a reduced noise requirement for that XMEGA so reduced digital (no USB).

There is some industrial use of USB; older methods are more likely and reliable (some concerns with USB though corrections for those concerns exist).

mojo-chan wrote:
Cost perhaps, ...
XMEGA E appear to challenge some tinyAVR other than on package size.

 

"Dare to be naïve." - Buckminster Fuller

Last Edited: Mon. Nov 2, 2015 - 01:00 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:

The XMEGA E ADC has some improvements (source impedance range, 16bit mode); therefore, a reduced noise requirement for that XMEGA so reduced digital (no USB).

 

Hmm... I'm not really convinced. They already say you need to disable other peripherals and sleep the CPU to make good 12 bit measurements with the ADC, so even on parts with USB it really has to be disabled if you want that.

 

I'd love to use an E5U in a number of projects, but have to user larger parts instead.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The semiconductor industry is market-driven, not dream- driven. 'If there is a market, they will build it', not 'If you build it, they will come'.
I have no doubt that if you put in an order for 5 million units of an Xmega with OTG, 'they will build it'.

"Experience is what enables you to recognise a mistake the second time you make it."

"Good judgement comes from experience.  Experience comes from bad judgement."

"Wisdom is always wont to arrive late, and to be a little approximate on first possession."

"When you hear hoofbeats, think horses, not unicorns."

"Fast.  Cheap.  Good.  Pick two."

"We see a lot of arses on handlebars around here." - [J Ekdahl]

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jalbinson wrote:
 the use of  larger SAM/ARM chips is unnecessary bloat.

Really?

 

USB (particularly Host) isn't exactly in the "lightweight" realm - is it...?

 

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

mojo-chan wrote:
... but that doesn't explain the lack of a USB device peripheral on E5. Cost perhaps, ...
An alternative is a CH340 (USB device and VUSB to 3.3V regulator) at about 0.35USD/1 for 100 or about 1USD/1.

The CH340T generates a DTR signal that, in the Arduino context, is used to reset the AVR; AVRDUDE can control DTR.

https://hackaday.com/tag/ch340/

www.olimex.com

BB-CH340T - Open Source Hardware Board

https://www.olimex.com/Products/Breadboarding/BB-CH340T/open-source-hardware

Olimex on Wordpress

New KiCAD OSHW PCB project – Getting rid of FTDI – BB-CH340T USB to Serial converter

26 Jun 2015

by Olimex

https://olimex.wordpress.com/2015/06/26/new-kicad-oshw-pcb-project-getting-rid-of-ftdi-bb-ch340t-usb-to-serial-converter/

...

Re-designing all our boards which were unfortunate to use FTDI chips.

CH340T is nice alternative to FTDI chips and have same features and even more, but cost exactly 1/10th of FTDI chip price.

...

Edits : added Olimex blog entry, DTR.

"Dare to be naïve." - Buckminster Fuller

Last Edited: Tue. Nov 3, 2015 - 08:37 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jalbinson wrote:
... believe it or not,  there is a real need to interface to USB slave devices from a small low power host.

...

as the use of  larger SAM/ARM chips is unnecessary bloat.

Bloat's popular wink

If by 'small low power host' you mean about 100mW then I see your problem.

Else, there's a near plethora of 1W ARM or x86 boards or SoM.

http://raspi.tv/2015/raspberry-pi2-power-and-performance-measurement

http://www.helios.de/web/EN/news/HELIOS-Edison.html

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Surely, there's equally a plethora of Cortex-M parts that can do USB Host at well under 100 mW ?

Top Tips:

  1. How to properly post source code - see: https://www.avrfreaks.net/comment... - also how to properly include images/pictures
  2. "Garbage" characters on a serial terminal are (almost?) invariably due to wrong baud rate - see: https://learn.sparkfun.com/tutorials/serial-communication
  3. Wrong baud rate is usually due to not running at the speed you thought; check by blinking a LED to see if you get the speed you expected
  4. Difference between a crystal, and a crystal oscillatorhttps://www.avrfreaks.net/comment...
  5. When your question is resolved, mark the solution: https://www.avrfreaks.net/comment...
  6. Beginner's "Getting Started" tips: https://www.avrfreaks.net/comment...
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

31 Cortex-M0+ via Atmel's MCU selector.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:

Surely, there's equally a plethora of Cortex-M parts that can do USB Host at well under 100 mW ?

 

Yes, my search above on Digikey found many sub $2 choices, all 32b but not all Cortex M.

 

If I search for USB Microcontroller (not OTG) the simpler USB-device parts are 8bit, with SiLabs (61c) and Microchip sub 99c, and also Freescale just above $1.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Two points -

 

1) Lack of USB Host in XMegas. USB host is substantially more complex than USB client. The state machine is much bigger and there is a lot more data that needs to be stored, probably in the form of registers. This would certainly raise the cost of a device in a family that has already been criticized for its relatively high cost compared to Cortex M devices..

 

2) Lack of USB, generally, in E5 series. You cannot simply turn off or disable USB. There is a LOT of activity that has to occur within considerable time constraints. I would guess that there is a fundamental conflict between the time to do an ADC conversion and the required USB response time.

 

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I was surprised that the E5 range didn't have USB at all.

I think of the Xmega E5 series as the "Tiny" of the Xmega lineup.

 

The Xmega lineup has the various series, with "A" being the big iron, with all of the features and lots of pins, (generally).

With the E5 being at the "low" end of the scale, I'm not surprised it doesn't have the USB OTG module.

 

Recall that the first Xmega to hit the market was an A series, (A128 ?), and it had a fair number of silicon bugs as it was.

The high speed ADC, the DAC, the DMA, the Event system, the encryption module, etc., were all new and improved over the Mega series.

At some point one has to draw the line on how much new and improved, (and untested), stuff one wants to try to integrate into a new product.

 

Note that the E5 series did add a new Timer/Counter module, and the programmable logic modules, (and another new DMA controller ?, I don't recall).

Those were added as the series has evolved over time, and hence are not found on the generally larger, more featured, A series.

 

Time will tell if the (larger) Xmega micro's get a USB OTG module, (or more likely yet another new Xmega series is created: A, B, C....E,  -> F <- ? ), or if the powers that be push designers to move to an ARM core.

 

All I can say is thank goodness the micros still contain the good old USART!

 

JC 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 2

No, but its quite possible to implement Host in an AT90SUB1287 in less than 1.2k to talk to a USB barcode scanner. (coded in assembler)

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

DocJC wrote:

Time will tell if the (larger) Xmega micro's get a USB OTG module, (or more likely yet another new Xmega series is created: A, B, C....E,  -> F <- ? ), or if the powers that be push designers to move to an ARM core.

 

Looking at new parts like the SAM09, I'd say that decision has already been made.

That pretty much outflanks the whole XMEGA series with ARM variants now.

(even a large chunk of AVR series, down to ATtiny are covered too... )

 

SAM09 has 32/48/96MHz clocking - Perhaps if the XMEGA had been Wide VCC parts, that may have given them more life.