There are few erratas in Engineering RevB(kits given out in GTT) that has been fixed in RevC
To see the differences you can see the documents listed here
S132V2.0.0 is officially being made and tested on RevC silicon. We recommend customers to use the final silicon for developing their products and not rely on an old silicon versions for development. However, the S132 v2.0 will work on “Engineering B” silicon but not in a reliable and a stable manner.
The Erratum 73 TIMER: Event lost which was present in RevB was fixed in RevC silicon. And S132v2.0.0 has removed the workaround needed for it as it is officially supporting only RevC silicon.
Does this mean that all of you who
have RevB cannot use S132v2.0.0 and
SDK11?
The answer for this is not straight forward. For all those who are making their development using RevB can still use it with the above mentioned versions of softdevice and SDK if the workaround for this Erratum 73 is applied as described in the Errata document for "Engineering B". If the workaround is not applied, the SoftDevice may occasionally trigger a fault (assert) due to missing events from the hardware when running on this chip revision. This assert will then have to be handled by the application’s assert handler. (This Erratum also affects the application itself in the same way.) Using this workaround will effect power efficiency.
Why was this workaround removed from
softdevice?
Note that the workaround impacts the power consumption, so we do recommend to use RevB only for development and if you need further power efficiency with latest releases then you need to use latest revision of silicon.
There are few erratas in Engineering RevB(kits given out in GTT) that has been fixed in RevC
To see the differences you can see the documents listed here
S132V2.0.0 is officially being made and tested on RevC silicon. We recommend customers to use the final silicon for developing their products and not rely on an old silicon versions for development. However, the S132 v2.0 will work on “Engineering B” silicon but not in a reliable and a stable manner.
The Erratum 73 TIMER: Event lost which was present in RevB was fixed in RevC silicon. And S132v2.0.0 has removed the workaround needed for it as it is officially supporting only RevC silicon.
Does this mean that all of you who
have RevB cannot use S132v2.0.0 and
SDK11?
The answer for this is not straight forward. For all those who are making their development using RevB can still use it with the above mentioned versions of softdevice and SDK if the workaround for this Erratum 73 is applied as described in the Errata document for "Engineering B". If the workaround is not applied, the SoftDevice may occasionally trigger a fault (assert) due to missing events from the hardware when running on this chip revision. This assert will then have to be handled by the application’s assert handler. (This Erratum also affects the application itself in the same way.) Using this workaround will effect power efficiency.
Why was this workaround removed from
softdevice?
Note that the workaround impacts the power consumption, so we do recommend to use RevB only for development and if you need further power efficiency with latest releases then you need to use latest revision of silicon.
Aryan can you just clarify a couple more points please
Engineering Rev A was on the PCA10036, the preview development kit
Engineering Rev B was on the PCA10040 0.9.? which was given out at the Global Tech Tour
So there's Engineering Rev C and there's Rev 1 which is the first production release (yay!). What versions of the dev kits have either of those two on them? Are there any differences between Engineering Rev C and Production Rev 1 at all or is it just a renaming?
I'd like to be sure the next PCA10040 I buy has Production Rev1 silicon on it, or Engineering C if that is really the exact same thing. Hence I need to ask Mouser, or whoever, very, very carefully, which revision of the PCA10040 they have in stock.
What versions of the dev kits have either of those two on them
I am not sure, I will have to check and comeback to you, if i forget please be gentle and remind me.
Only two minor improvement as far as i can tell in Rev1
fixed. 53. “Using SCB.AIRCR.VECTRESET may get the chip into an undefined state”
the calibration and tuning values in FICR might have been optimized for rev1, but the silicon is identical in other aspects other than these two points.
Hi RK,
There is IC Revision overview where you see the packet variant and build code. Ask your supplier if the DK they have are with ICs with Rev1 build codes on the chip. But I do no think anyone has them yet.
PCA10040 v0.9 has the rev B
PCA10040 v1.0.1 is RevC
donno which one is Rev1
Thanks - will. I see Mouser is out anyway right now so hopefully they are getting new kits in with new revs. I'll ask a couple of other guys I use too, even though my GTT kit is working just fine.
We ordered 4 DKs (digikey) late on Tuesday and they came in today marked as PCA10040 v1.1.0 which according to ([https://devzone.nordicsemi.com/question/71155/buy-pca10040-with-latest-ic-revision]) is the latest chip revision . Each Dev Kit shipped with a strip of 5 additional N52832-QFAAB0-1606AA SoCs.
