PIC32MZ Errata

PIC32MZ Errata

Postby floatingpoint » Fri Jul 11, 2014 1:20 am

Anyone have a view as to the future of the PIC32MZ?

Errata is not a good look:

http://ww1.microchip.com/downloads/en/D ... 0588D_.pdf

Oscillator POSC 41. A crystal oscillator cannot be used as an input to the Primary
Oscillator (OSC1/OSC2 pins).

ADC — 44. ADC module does not meet published specifications
This appears to mean that 12bit has now become 10bit

Then there was word that USB module will not run at high speed??

Not to mention the unavailability of the chips themselves...

FP
floatingpoint
 
Posts: 4
Joined: Fri Jul 11, 2014 12:39 am
PIC experience: Professional 5+ years with MCHP products

Re: PIC32MZ Errata

Postby Tom Maier » Fri Jul 11, 2014 12:53 pm

I think microchip is using all their available chips for use in their demo boards.

http://www.digikey.com/product-search/e ... ?k=pic32mz

They probably won't release significant quantities of single chips until they get the major bugs knocked out, or work-a-rounds found.

Many years ago I worked for a company that designed some new machines that grossly failed the field beta testing, and they ignored that and went into full production. It created a year of excellent sales, but the financial damage over the next three years was heavy. The high failure rate also smeared the company name. I called it "manufacturing liabilities at high volume".
User avatar
Tom Maier
Verified identity
 
Posts: 179
Joined: Mon May 26, 2014 2:37 pm
PIC experience: Professional 5+ years with MCHP products

Re: PIC32MZ Errata

Postby drh » Fri Jul 11, 2014 3:27 pm

It should be an "interesting" Masters Conference this year, if the right "MZ" questions are asked.
User avatar
drh
Verified identity
 
Posts: 56
Joined: Tue May 27, 2014 3:31 pm
Location: Hemet, Calif.
PIC experience: Professional 5+ years with MCHP products

Re: PIC32MZ Errata

Postby floatingpoint » Fri Jul 11, 2014 11:32 pm

Tom Maier wrote:I think microchip is using all their available chips for use in their demo boards.

http://www.digikey.com/product-search/e ... ?k=pic32mz

They probably won't release significant quantities of single chips until they get the major bugs knocked out, or work-a-rounds found.

Many years ago I worked for a company that designed some new machines that grossly failed the field beta testing, and they ignored that and went into full production. It created a year of excellent sales, but the financial damage over the next three years was heavy. The high failure rate also smeared the company name. I called it "manufacturing liabilities at high volume".


Yes, we all expect some teething problems with new product releases.
However the Rev3 and Rev4 errata appears to be little changed.
Without much knowledge of the silicon debugging process, I'm interested to know what percentage of the major bugs are likely to go in Rev5?

Also the 12 bit to 10 bit change for A/D removes one major attraction for me to use this chip in future.
I can't decide if this is a change to the core resolution of this module or previous information was stating 12 bit resolution only if oversampling was used.

FP
floatingpoint
 
Posts: 4
Joined: Fri Jul 11, 2014 12:39 am
PIC experience: Professional 5+ years with MCHP products

Re: PIC32MZ Errata

Postby Tom Maier » Sat Jul 12, 2014 12:21 pm

Most likely the ADC is degraded from 12 bit to 10 bit due to imperfections in the analog section. It's easy and low failure to put in the digital bits, and the highest failure item is the analog performance of the comparator. The measurement errors probably are too great to call it a 12 bit adc.

My guess...
User avatar
Tom Maier
Verified identity
 
Posts: 179
Joined: Mon May 26, 2014 2:37 pm
PIC experience: Professional 5+ years with MCHP products

Re: PIC32MZ Errata

Postby floatingpoint » Mon Jul 14, 2014 9:56 pm

Tom Maier wrote:Most likely the ADC is degraded from 12 bit to 10 bit due to imperfections in the analog section. It's easy and low failure to put in the digital bits, and the highest failure item is the analog performance of the comparator. The measurement errors probably are too great to call it a 12 bit adc.

My guess...


I expressed these concerns to our local MChip Office and received this reply:

MChip wrote:Regarding the PIC32MZ:

· We are about to release samples of revision A5. Here’s the schedule:

o Limited samples: August

o Production release: September

· I have been told if you use an external reference it can help improve the accuracy of the PIC32MZ ADC. Mark, our FAE, can help if you have questions on this.

· We are working on a new all-layer redesign which will be released as PIC32MZxxxA. This will not only include a new ADC and (hopefully) fix all known errata, but will also include an updated SPI that will improve the speed to 50 Mbps, allow us to introduce E-temp devices, and add the missing FPU. These parts are expected to sample in March 2015 and release to production in June 2015. The ADC will be 12-bit although I don’t have any speed specs at the moment.


Some good news (including an FPU) albeit a year away...

FP
floatingpoint
 
Posts: 4
Joined: Fri Jul 11, 2014 12:39 am
PIC experience: Professional 5+ years with MCHP products

Re: PIC32MZ Errata

Postby DominusDRR » Mon May 04, 2015 4:59 pm

Does anyone have any new news about MZ version A?

:?: :?:
User avatar
DominusDRR
 
Posts: 1
Joined: Mon May 04, 2015 4:53 pm
PIC experience: Professional 5+ years with MCHP products


Return to PIC32 topics

Who is online

Users browsing this forum: No registered users and 2 guests