I have been using ATtiny85 chips for years with zero problem. Last week i bought a few more and now these do not bootload or allow me to upload anything. Does anyone know if something changed with the newer version of these chips. ( I have tried several of the new ones all give same results)
I am using Arduino as ISP. I have been using the same setup for years with the older chips. Not until this week has there been an issue.
these do not bootload
Do they have a bootloader loaded ?
clawson 
do not bootload
I am using Arduino as ISP.
So all things equal i am able to burn bootloader or sketch to one chip but not the other (w/o changing anything) I attached logs.
The older chips i have all work (see photo) the newer ones none work. I am using the exact same setup for both.
Can you post a picture of your programming setup?
I have been using ATtiny85 chips for years with zero problem.
sounds like an intermittent (broken) wire, sometimes it works, sometimes it does not.
Jim
clawson Until the signature can be read your are not "in contact" with the chip. As I say this is usually a wiring or a clocking issue so those are the things you need to explore.
Diagnostic tools like scopes (esp DSOs, logic analysers and so on may help in detecting which lines are not behaving as expected.
Sometimes where avrdude is being invoked it can help to add a number of -v to the command line to get info about what is/isn't happening but if even the signature cannot be read it basically means "not talking"
Reading | ################################################## | 100% 0.03s
avrdude: Device signature = 0x000000 (retrying)
Reading | ################################################## | 100% 0.03s
avrdude: Device signature = 0x000000 (retrying)
Reading | ################################################## | 100% 0.03s
avrdude: Device signature = 0x000000
avrdude: Yikes! Invalid device signature.
Double check connections and try again, or use -F to override
this check.So it's failing to read the device signature.
Where did you buy them? Is it a reputable supplier? Have you contacted them about this?
post a picture of your programming setup
+1
See Tip #1 for instructions ...
So it's failing to read the device signature.
Where did you buy them? Is it a reputable supplier? Have you contacted them about this?
LOL reputable.. unlikely i got them from aliexpress so who knows how or where these were made..:-)
I have used this header for a long time and never had any issues. (see pic)
LOL reputable.. unlikely i got them from aliexpress
So it's a chip with an fake Atmel logo then....
You didn't follow the instructions for how to post the picture, then.
So we can all see it in context:
So you're trying to program a surface-mount chip in a DIP socket?!
As Jim suggested, that sounds like a recipe for flaky connections ...
clawson
Dare one ask what's going on here:
That pin looks like it's got another 2..3mm to go before it's firmly connected ?!?
Sorry I thought thats what you wanted Here is another one.
I am tying Surface mount. this header And i get what your suggesting about wiring but I can use this exact setup for the working chips no issues at all. Infact I just did it again successfully flashed 4 chips (the older ones) and the others failed. However i did get a new error
avrdude: Device signature = 0x1e9005
avrdude: Expected signature for ATtiny85 is 1E 93 0B
Double check chip, or use -F to override this check.
looks like from what i googled thats a t12?
david.prentice Working chips: made in 2013
Non-working: made in 2017
It is now 2019. I would expect to see datestamps for 2019.
I am gobsmacked by your setup. Would it not be easier to just buy a USBASP (with 3x2 header) ?
like this
Have you tried old chips in your clam-shell adapter?
The clam-shell might have dirty contacts.
Your "wiring" from the Uno might have a broken wire.
Life is much safer when you can plug a ready-made ribbon onto soldered 3x2 headers.
When the ribbon gets old or worn, buy a new ribbon.
David.
I didn't have completely ready when i took a picture. So yes that pin is hanging. I can assure you it is not when i use it.
Sorry you're "gobsmaked" but setup works perfectly well I have never had an issue. yes I agree its not pretty. So to answer you yes i have tried this exact setup with the older chips and they work just fine. I went back and tested every chip with this setup (making sure everything is snug and tight) I separated out the errors. And found that one set of chips fails differently. see picture
One set errors with 0x000000 one erros with 0x1e9005
PS clam shell is brand new
see picture
Did I not mention Tip #1 - which shows you how to properly post pictures ?
sorry i dont use forums much and i just saw add attachments.. I have determined that the issue is a batch of fake chips from china.
Lesson here is order from reputable source. Picture below are both fake chips hard to tell the real from fake.
gchapman ... or where these were made..:-)
Country-of-Origin (CoO) is on package's bottom by the injection molding machine.
After 1-Sept'17, only CoO is on the bottom.
https://www.microchip.com/mymicrochip/Reports.aspx?type=cpn&filter=ATtiny85
Product Change Notification - GBNG-15KQFZ896 - 11 Jul 2017 - CCB 2971 Final Notice: Implement marking changes for virtually all Atmel products (first attachment)
http://www.lingsen.com.tw/webe/html/about/about.aspx
edit :
Location: Chandler AZ
Lesson here is order from reputable source.
LOL - hence my question in #7 !
I have determined that the issue is a batch of fake chips from china.
Am I allowed to pay them with fake money? Seems equal value for equal parts.
I hope you didn't make the mistake I made before by accidentally set the chip to use the external clock.
If you suspected to made that mistake, try putting the crystal oscillator between Pin 2 and Pin 3 and reupload the bootloader with Internal Oscillator setting selected.
Good Luck!
clawson , try putting the crystal oscillator between Pin 2 and Pin 3 and reupload the bootloader with Internal Oscillator setting selected.
https://www.avrfreaks.net/forum/...
david.prentice I remain sceptical. I can believe that a punter has done something silly e.g. bad wiring, inappropriate fuse setting, ...
I can't believe that anyone would try to sell counterfeit chips.
After all, the buyer is unlikely to purchase 10000 in the first order. Any smaller quantity is "not worth the scam" from the vendor.
Whatever the size of the purchase, the buyer is not going to repeat her order.
It is possible that a batch of pre-programmed chips appear on the Ebay / AliExpress market. e.g. surplus inventory from an assembly house with fuses set for an external clock.
So it is wise to examine your "broken" chips with a proper programmer (and good wiring).
David.
I can't believe that anyone would try to sell counterfeit chips.
And yet, clearly, they do!
Whether it is the case here remains questionable - but counterfeit chips certainly are A Thing.
https://www.sparkfun.com/news/364
Remember the fake FTDI stuff a few years ago ?
And there's the well-known fake nrf24l01 chips - which can be identified by an operational bug; eg,
https://www.avrfreaks.net/commen...
david.prentice Think about it. You would have to set up a machine to scrub off the existing laser printing. Then set up another machine to print the fake numbers.
Practical for $10 DIP-40 chips. Sell through Ebay. Most customers do not complain.
Not practical to do "by hand". $0.50 SMD chip in reel packaging. And you still don't sell that many through Ebay.
Yes, if you work in a Semiconductor Fab and can access the printing and packaging.
I am sure that Ebay vendors have got better ways to scam the public. (and actually make money)
David.
Distant cousins there's a limited supply.
I have determined that the issue is a batch of fake chips from china.
Before tossing the batch into a bin, try HVSP on one or more of them.
Do those which report 0x1e9005 do so consistently?
|
"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] |
 |
hey guys thanks for the feed back. I tested each chip with my "janky" setup and they consistently report 0x1e9005 or 0x000000 and the older ones i have all work as expected. This leads me to think its NOT my wiring or setup but rather that the chips are bogus. I did report this to ali express and to their credit have issued a refund. So like i said lesson learned i ordered the new owns from Digi-Key and hopefully they won't be an issue.
I have a mac and have not found a great tool to flash them other than arduino any suggestions
I have a mac and have not found a great tool to flash them other than arduino any suggestions
MPLABx perhaps!
This leads me to think ... that the chips are bogus.
But what David ans joeymorin are saying is that it is more likely that the chips are genuine, but have been previously programmed - leaving them in an incompatible state with your setup.
In which case you should be able to "recover" them as described ...
david.prentice And you can even add a "clock"signal to your clamshell.
It is possible but unlikely that RSTDISBL fuse has been programmed.
It is quite likely that EXTCLOCK or EXTXTAL has been programmed.
Did your SMD chips come in sealed tape?
Rigging up HVSP is relatively easy. You would then get an accurate answer. e.g. non-virgin Fuses and/or non-virgin Flash.
David.
In which case you should be able to "recover" them as described ...
|
"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] |
|
