Redesigned betemcu.cn usbasp - Request for comments

[rperkins]

The manufacturer of the betemcu.cn usbasp has redesigned their board . It's now only populated on one side. That's the same thing HK recently did. The board still shows supporting 3.3v programming. I see what could be a VR, albeit smaller than the current one.

What do you guys think ?

UPDATE- I was told these have been out a month. I'd love to hear anyones experience using this revised version with a 9x/9xr

The salesman also mentioned the clamshell cases for them. They cost nearly as much as the programmers. My thought is 'who needs it' but I'd get some if there is an interest

[jhsa]

shrink tube them?

[ShowMaster]

If the programmer is dedicated I used clear shrink tubing but a clam shell case would be easier to get at jumpers and give better component protection .
How much $?

For my 9x cases
I'm thinking of actually removing the USB and 10 pin connector from your programmer and adding a sparkfun USB breakout board to it. Then mount the programmer inside the case and the modified sparkfun USB inside the battery compartment.
The stripped down programmer should be easy to mount inside the case out of the way and have a small footprint.

I like SkyNorth's ps2 connector setup for access to programming and the programmer can be internal or external.

The USBasp is inexpensive enough that I don't mind dedicating it to the case internally. The 100 ohm resistors offer some isolation from being on the m64/128 programming lime full time.

This is what I've done for my ver B skyboard tx case to connect to the USB port.

Sent from my iPod touch using Tapatalk.

[rperkins]

The cases are $3 $1 when ordered with a programmer. Not worth it to me butNot bad for a dollar and it does dress it up and hides the circuit for people with those fears/concerns.
Yea shrink wrap would be more my speed.
Lots of people sell the clam cases so there must be a demand. I'm just trying to gauge to what extent in our little corner of the world

http://m.aliexpress.com/item/900789283. ... sitedetail

[rperkins]

I placed a small order for the new boards to test them out. It' not the usbasp units I am concerned about. It's how they work in a 9x/9xr considering the circuitry on those boards. The 9x/9xr uses the programming pins for other duties when it is not being programmed.

I did some more research on their revisions
betemcu.cn 2011 - What we all are familiar with. 3.3 and 5v operation . dual sided
betemcu.cn 2012 - havent seen around here - 5v only - single sided - usually offered with light blue case
betemcu.cn 2013 - the board pictured in this thread - 3.3 and 5v operation - single sided - will work with light blue case

[Kilrah]

Are you flashing the open source FW yourself, or have you tried to talk them into flashing it on all their units themselves? Wonder about the reason for using the "weird" FW...

[rperkins]

I flash them myself using my 'usbasp cloner' which is made out of a ...... usbasp
http://openrcforums.com/forum/viewtopic.php?f=85&t=3234

I can only guess why they use their own firmware
1. Truly believe the 'autospeed' feature provides value. And it did before the -B option to avrdude. still does to some people. I'd like avrude to incorporate an autospeed feature. But I cant say too much cause I havent worked towards it myself.
2. Afraid they may be perceived as 'owing' the open source community something if they use that firmware.
3.

Why I use the open source firmware
1. The source
2. usbasp is an open hardware design. putting closed source firmware on it just feels wrong
3. complete control over the speed with the -B option
4. No nagging warning about 'cannot set sck period. please check for usbasp firmware update.'
5. differentiate my product from the competition.

No I havent asked them to put the firmware on there. Thought about it. I dont sell enough nor have developed the relationship to that point at this time. I've minimized the time it takes for me to do it with the 'usbasp cloner' .

[rperkins]

after looking at the picture best I can it appears to be basically the same thing except:
1. 3.3 volt VR is smaller
2. original model has 4 caps on 3.3 v VR circuit ( 2 on input, 2 on output ) . new model only has 2 caps

and there probably are traces on the back side. cant see them in picture but can see the vias

[rperkins]

Hi Guys,
I got five of these units in for testing and successfully used it with a 9xr. I will be transitioning to them as my current supply is depleted.
They use 100 ohm resistor packs for current limiting on the data lines and an M8 MCU which has 8KB of flash. The HK usbasp that was recently pulled was using M48PA , which only has 4KB of flash memory. The latest versions of USBASP firmware will not fit within 4KB of flash, so I am glad betemcu.cn continued using the M8.

Also ordered some of the cases as an option for people who are interested. Was given a sample and they clip together sturdily and finish out the USBASP nicely. Expecting to price them around $1 when shipped with a usbasp

[arthur92710]

I've registered just to comment on the updated programmer.

I have both the first version(2011) and this new one(2013).

What I have immediately noticed is that 2013 has an issue with voltage on the output. I'm using them for AVR microchip programing and powering my projects.

Normally I would use the first ground pin next to the voltage pin on the 2011 board. When I used the new board on a working project it would only show ~1-2v across vcc and ground.

After some investigation, I noticed that some of the ground pins on the 2013 board have a high resistance to the usb ground connector, around 2000ohms.

I switched to one of the low resistance grounds and everything worked as it should.

So, if you have any issues, check the ground resistance.

[rperkins]

Thanks for the information. I've used them with the included 10 pin cable and sometimes the 10 to 6 pin adapter, but never as you mentioned.

You mention the pin next to the VCC. That would be pin 4. Yes the atmel docs and most of the images posted online of the pinouts show that pin as going to ground. Thomas Fischl, the designer of usbasp, set aside pins 4 and 6 for a serial com port. The idea was that USBASP could program your target device and also provide a com port for debugging. However this feature was never fully implemented.

You can take a look at the original usbasp schematic here http://www.fischl.de/usbasp/. It shows pin 4, the one you were using as ground, as running through a 1k resistor back to the PD1 port of the MCU. Additionally, pin 6 runs back to pin PD0. Also if you look at the 'features' listed at the same URL provided above, The last item mentioned is 'Planned: serial interface to target (e.g. for debugging).'

Pins 8 and 10 are listed as ground on the Fischl schematic. Really this is good news if the new version of the Usbasp brings those ports out to those pins. If you decide to repurpose a usbasp for another task, you could utilize those port pins easily. But I admit it makes it difficult when their earlier version did not follow that same implementation.

I'll do some checking tonight. Been wanting to get an updated schematic but hadnt took the time to look that close at the new revision.

I've been shipping this new version for a couple weeks now and havent had any issues reported. Also shipping the new 10 to 6 pin adapter as posted here http://openrcforums.com/forum/viewtopic ... 951#p53593. The new clamshell mentioned in this thread and pictured Here is fairly popular (~40%) and only add a dollar to the cost when purchased with a programmer.

Thanks for taking the time to post this. I like hearing about how these little devices are working out for people. Oh and welcome to the forum

[rperkins]

Ok I looked at the new board and those pins are not ground, they are used for the unfinished serial port. If you use the 10 to 6 pin adapter it does short those alternate pins to ground. Found this and it appears to be this recent revision. On the lower right are pins 8 and 10 which are ground. 4 and 6 work their way up through some resistors to the PD1/PD0 port pins. VCC is on pin 2 and gets it's power from the other side of the board,

[arthur92710]

So the hardware is there, but not the software. It would be a nice feature to have.

[rperkins]

that's correct. One thing I noticed is that the virtual usb, http://www.obdev.at/products/vusb/index.html , implementation used in usbasp supports multiple endpoints, but multiple endpoints violates the low speed usb protocol. Maybe that is important maybe not. Maybe it would be something that could be implemented in a bootloader and only one function was active at a time. Right now it's all just beyond my grasp.