Ian Lesnet Works Around the RP2350's Erratum E9 to "Fix the Six" — with Two Resistor Packs
If you've got a first-issue Bus Pirate 5XL or Bus Pirate 6, two resistor packs and a quick solder job should get them working as-expected.
Dangerous Prototypes' Ian Lesnet has come up with a way to get the Bus Pirate 5XL and Bus Pirate 6, previously put on hold as a result of a hardware flaw in Raspberry Pi's new RP2350 microcontroller, back up and running: add a couple of resistor arrays.
"Fix Your Six! The new RP2350-based Bus Pirates need an extra external pull-down resistor to fix silicon bug E9," Lesnet explains. "We need to replace the pull-down resistors on the RP2350 pins that control the Bus Pirate IO pins: RN302, RN307. We need two replacement resistors: 4.7k resistor array (4.7k-8k should all work), 4×0402 package (also called 0804 or 2010M). Convex type (pads on the end of the leads, not between the leads)."
The "silicon bug E9" to which Lesnet refers is erratum RP2350-E9, a hardware flaw in Raspberry Pi's recently-launched quad-core dual-architecture RP2350 microcontroller caused, the company says, by IP received from an outside vendor and included in the chip's design. It exhibits as a problem when using general-purpose input/output (GPIO) pins as an input, causing them to "latch" at around 2.15V — something the company had previously said only happened while using the chip's internal pull-down resistors, but later admitted could be triggered even without these active.
The flaw was enough for Lesnet and others to pause or outright cancel planned RP2350 projects pending revised silicon or at least more clarity on the issue, with Lesnet warning that RP2350-based Bus Pirate 5XL and Bus Pirate 6 hardware — designed as RP2350-powered successors to the RP2040-driven Bus Pirate 5 debugging tool — shipped before the problem was discovered "are probably DOA [Dead On Arrival]."
For those who had picked up a "DOA" Bus Pirate 5XL or Bus Pirate 6, then, Lesnet's discovery of a workaround for the flaw will be welcomed: the soldering of two resistor packs, which act as an external pull-down strong enough to resolve the latching behavior and have the GPIO pins work as expected. Better still, it's a relatively straightforward fix: "Instead of removing the existing 100k pull-downs," Lesnet explains, "we can solder the new resistor array on top."
Full details on the fix are available in Lesnet's Mastodon thread; those who have a Bus Pirate 5XL or Bus Pirate 6 and who perform the modification can send Lesnet a picture of the fix to receive an exclusive "I Fixed My Six" sticker.
Freelance journalist, technical author, hacker, tinkerer, erstwhile sysadmin. For hire: freelance@halfacree.co.uk.