Where to buy a RP2350 chip currently?

[SiliconWizard]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

[phil from seattle]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

'Cause they can...   At Q100, it drops off to $1.40.

[phil from seattle]

I had them build a run of 5 boards based on the RP2350B.  Got them 2 days ago. FWIW, they work.  And, no, I didn't use consignment.

working on a few projects with it as well, did you went with the recommended layout for the switching regulator, i am doing a similar one as but on a 4 layer board.

I did a 2 layer design and it works quite well.  However, mostly digital and I don't have any signals above 400 kHz.

I want to replace the Abracon parts but they are still towing the RPi beta nonesense about using them.  I'll try pushing it past them...

[SiliconWizard]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

'Cause they can...   At Q100, it drops off to $1.40.

Yes but it's overall twice the price of the RP2040. I didn't think there would be such a difference between the two, but I was probably mistaken or misled by initial announcements from the RPi?

[phil from seattle]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

'Cause they can...   At Q100, it drops off to $1.40.

Yes but it's overall twice the price of the RP2040. I didn't think there would be such a difference between the two, but I was probably mistaken or misled by initial announcements from the RPi?

That is the price for the RP2350B. The RP2350A which is largely equivalent to the RP2040 is a bit cheaper - $1.20 Q100. Still a lot more than $0.70 for the RP2040 Q100.

To consider costs, you probably should factor in the Abracon inductor and crystal because JLCPCB is saying you need to use those.  That's another $0.54 Q100. Maybe the 2350 will drop in price when their "beta" finishes.

[SiliconWizard]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

'Cause they can...   At Q100, it drops off to $1.40.

Yes but it's overall twice the price of the RP2040. I didn't think there would be such a difference between the two, but I was probably mistaken or misled by initial announcements from the RPi?

That is the price for the RP2350B. The RP2350A which is largely equivalent to the RP2040 is a bit cheaper - $1.20 Q100. Still a lot more than $0.70 for the RP2040 Q100.

To consider costs, you probably should factor in the Abracon inductor and crystal because JLCPCB is saying you need to use those.  That's another $0.54 Q100. Maybe the 2350 will drop in price when their "beta" finishes.

That's right, yes we need to factor in the external components required (for the inductor, for those wary of not routing this properly and run into the known issue, they may want to add an external regulator which will further add to the cost).
Yes, maybe prices will go down when it's in full production mode. It's still not clear at this point what the production plan is here, given how "slow" they are to release it officially on a large scale.

[phil from seattle]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

'Cause they can...   At Q100, it drops off to $1.40.

Yes but it's overall twice the price of the RP2040. I didn't think there would be such a difference between the two, but I was probably mistaken or misled by initial announcements from the RPi?

That is the price for the RP2350B. The RP2350A which is largely equivalent to the RP2040 is a bit cheaper - $1.20 Q100. Still a lot more than $0.70 for the RP2040 Q100.

To consider costs, you probably should factor in the Abracon inductor and crystal because JLCPCB is saying you need to use those.  That's another $0.54 Q100. Maybe the 2350 will drop in price when their "beta" finishes.

That's right, yes we need to factor in the external components required (for the inductor, for those wary of not routing this properly and run into the known issue, they may want to add an external regulator which will further add to the cost).
Yes, maybe prices will go down when it's in full production mode. It's still not clear at this point what the production plan is here, given how "slow" they are to release it officially on a large scale.

Actually, the design I have for an external 1.1V LDO (TLV74311, $0.035 q100) is quite a bit cheaper than the Abracon inductor ($.28 q100). It takes up about the same amount of board space, too - U19 in the picture. Add a few cents for the capacitors and call it $0.05).  I'd be way happy to find an inductor with a "polarity" marking for $0.05 but that doesn't seem to be in the cards.

Yeah, their slow walking the chip into production seems strange. Might be related to the security problems they have been shown.

[Bruno_PT]

JLCPCB appears to have > 6k in stock at the moment (for assembly). At > $2/1, I find it pricier than expected though?

'Cause they can...   At Q100, it drops off to $1.40.

Yes but it's overall twice the price of the RP2040. I didn't think there would be such a difference between the two, but I was probably mistaken or misled by initial announcements from the RPi?

That is the price for the RP2350B. The RP2350A which is largely equivalent to the RP2040 is a bit cheaper - $1.20 Q100. Still a lot more than $0.70 for the RP2040 Q100.

To consider costs, you probably should factor in the Abracon inductor and crystal because JLCPCB is saying you need to use those.  That's another $0.54 Q100. Maybe the 2350 will drop in price when their "beta" finishes.

That's right, yes we need to factor in the external components required (for the inductor, for those wary of not routing this properly and run into the known issue, they may want to add an external regulator which will further add to the cost).
Yes, maybe prices will go down when it's in full production mode. It's still not clear at this point what the production plan is here, given how "slow" they are to release it officially on a large scale.

Actually, the design I have for an external 1.1V LDO (TLV74311, $0.035 q100) is quite a bit cheaper than the Abracon inductor ($.28 q100). It takes up about the same amount of board space, too - U19 in the picture. Add a few cents for the capacitors and call it $0.05).  I'd be way happy to find an inductor with a "polarity" marking for $0.05 but that doesn't seem to be in the cards.

Yeah, their slow walking the chip into production seems strange. Might be related to the security problems they have been shown.

really hope they aren't wasting our time just because of "security issues", specially as the ones i seen require crazy commitment, if its a respin, then it would be warranted, but bottom line for such a "open" company, its been 8 months now, end of year came and gone, and zero comms.

[Bruno_PT]

Actually, the design I have for an external 1.1V LDO (TLV74311, $0.035 q100) is quite a bit cheaper than the Abracon inductor ($.28 q100). It takes up about the same amount of board space, too - U19 in the picture. Add a few cents for the capacitors and call it $0.05).  I'd be way happy to find an inductor with a "polarity" marking for $0.05 but that doesn't seem to be in the cards.

Yeah, their slow walking the chip into production seems strange. Might be related to the security problems they have been shown.

btw since we're here, don't this "voltage rings" layout that they have on the reference design seem to be a horrible idea for EMI?, i get that only the 3,3v wirebonds reach the edges of the chip, but still having the signals travel over 2 different "planes"/ polygons can't be good, or am i missing something?

[westfw]

hope they aren't wasting our time just because of "security issues"

I'm inclined to agree, but adding enough "code protection" to make the RP usable for proprietary applications WAS (AFAIK) one of the major changes in the new chips, so they might see it differently.

[phil from seattle]

btw since we're here, don't this "voltage rings" layout that they have on the reference design seem to be a horrible idea for EMI?, i get that only the 3,3v wirebonds reach the edges of the chip, but still having the signals travel over 2 different "planes"/ polygons can't be good, or am i missing something?

I don't know the answer to that.  The whole inductor design with the on board buck regulator is a bit opaque to me. And, there is quite a bit of discussion here and other places about the lack of wisdom in the design.  The requirement for a polarized inductor definitely feeds the fire.

really hope they aren't wasting our time just because of "security issues", specially as the ones i seen require crazy commitment, if its a respin, then it would be warranted, but bottom line for such a "open" company, its been 8 months now, end of year came and gone, and zero comms.

You and me both.  Though one of the complaints about the RP2040 was lack of security.  Perhaps they are looking to broaden the market for the 2350.  But, I am pretty sure they are doing a quick turn of the design, the pull-down problem is a pretty serious defect. Though, it doesn't affect my application.

[SiliconWizard]

If you look at the list of errata for this chip, there are quite a few other issues as well, some that would be worth fixing. Don't know what the plan is.