 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Cpu speed clock Olimex rp2350b XL and Weact rp2350b
| Author | Message | ||||
| guma1975 Newbie  Joined: 13/01/2023 Location: PolandPosts: 16 |
Hello. I bought a few olimex modules.  Unfortunately the maximum CPU clock is 240MHz option cpuspeed 240000 above it hangs or degrades to a lower clock. However, I tested the Weact module from this link successfully at 378MHz. Weact rp2350b The memories differ only in capacity and are from the same family and company. I only wonder if it is not a problem of the length of the paths leading to this memory. In Olimex, pin 2 and 3 of this memory is shorter than 5 and 7. In Weact, there are visible zigzags of PCB paths on pins 2 and 3. Could this be the reason that Olimex does not work above 240MHz? Olimex memory...  Weact memory...  And photos of the rp2350b itself, the markings are the same. The only other ones are below the markings, maybe it's just the series?   Or maybe the core voltage is too low and maybe it can be increased a bit? Is it feasible? I wonder what you all think about this. Best regards. |
||||
| dddns Senior Member  Joined: 20/09/2024 Location: GermanyPosts: 259 |
Many thanks for this report! |
||||
Mixtel90 Guru  Joined: 05/10/2019 Location: United KingdomPosts: 7499 |
Basically, if you buy something that's specified for 150MHz then it doesn't work at 151MHz it's within spec and you have no argument. :) TBH I doubt if there's much you can do on a ready built module to change things unless you are very good at SMD work. They'll have been designed for profit, not for overclocking so as long as they meet the official Raspberry Pi spec that's fine. I'm interested in the fact that Raspberry Pi haven't released a module for the RP2350B yet - maybe they know something about it that we don't (or at least shouldn't!). Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| guma1975 Newbie Joined: 13/01/2023 Location: PolandPosts: 16 |
Thanks for the answer. I can handle smd under a microscope, it's easy to operate with such a small thing. As for overclocking the clock, I have no complaints, I'm just curious if something can be done with it, even burn it... hehe. (release smoke...)  |
||||
javavi Guru Joined: 01/10/2023 Location: UkrainePosts: 436 |
Delete Edited 2025-03-16 02:20 by javavi |
||||
| Mixtel90 Guru Joined: 05/10/2019 Location: United KingdomPosts: 7499 |
Peter used a linear reg and made it adjustable to get overclocking. Maybe have a look at how he did it? Mick Zilog Inside! nascom.info for Nascom & Gemini Preliminary MMBasic docs & my PCB designs |
||||
| guma1975 Newbie Joined: 13/01/2023 Location: PolandPosts: 16 |
I soldered to weact psram from aliexpress works at 378MHz. Additionally there was a place for a capacitor I soldered 100nF.  |
||||
Sasquatch Guru Joined: 08/05/2020 Location: United StatesPosts: 376 |
guma1975: Thank you for providing this information about the Olimex and WeAct modules. Most of the RP2350B modules I have tested do not work at 378Mhz Some seem to only work at high speed with the PSRAM removed. It is good to know there are a few modules that will work at 378Mhz. I suspect that we may see some variation with individual RP2350 devices. -Carl -Carl |
||||
| matherp Guru Joined: 11/12/2012 Location: United KingdomPosts: 10066 |
Mine do, with or without the PSRAM  |
||||
| Sasquatch Guru Joined: 08/05/2020 Location: United StatesPosts: 376 |
Yes! Your design with the adjustable linear regulator for the core voltage may be the key to reliable overclocking success! I am waiting for the RP2350B HDMIUSB reference design to be released. Do you feel like the RP2350B DIL reference design is complete? Or should I wait for the final, final, final tweaks -Carl |
||||
| WhiteWizzard Guru Joined: 05/04/2013 Location: United KingdomPosts: 2927 |
I have had several of Peter’s 64 pin DIL modules on test (with various versions of the 2350 firmware) and they all seem stable. Have also supplied quite a few to UK members and not heard of any issues (perhaps some of those people can feedback here if experiencing anything). For a minimalistic 2350B design realising all available GPIOs I don’t see that there is anything to wait for |
||||
| guma1975 Newbie Joined: 13/01/2023 Location: PolandPosts: 16 |
I replaced the 25Q16 memory in Olimex with a 25Q128 from AliExpress and the speed at which the board works increased to 330 MHz, above that it doesn't work. That's a small success. |
||||
| PhenixRising Guru Joined: 07/11/2023 Location: United KingdomPosts: 1139 |
Did you test after removing the 25Q16 before adding the 25Q128? |
||||
| guma1975 Newbie Joined: 13/01/2023 Location: PolandPosts: 16 |
No. I think pico won't start without this memory. From 25q16 there was only about 36kb for the program by issuing the files command. From 25q16 max speed 240MHz from 25q128 max speed 330MHz. |
||||
| guma1975 Newbie Joined: 13/01/2023 Location: PolandPosts: 16 |
Hello again, encouraged by the fact that in the olimex pico2 rp2350 XL, replacing the flash with 25q128 improved the state of affairs regarding the operating speed and the size of the disk A: I decided to add the same psram that I soldered to weact. Despite adding a pull-up resistor R16=10k and C26=100nF according to the diagram and soldering the psram, everything hangs after issuing the option psram pin gp8 command. Unfortunately, this pcb probably won't work with psram. I changed the psram to the second one again, maybe the first one was damaged. Unfortunately, it didn't work either. I checked if gp8 pin6 rp2350 is connected to pin1 psram - it's ok. The power supply is 3.28V but on pin 1 psram is sometimes 2.65V, I could check with an oscilloscope, but I didn't. If I sleep on it, maybe I'll figure something out. One thing worries me. All pins for flash memory and psram are brought out to the main bus, i.e. gold pins or whatever it's called. Maybe the lines are getting too long and the parasitic capacitances are doing their thing. But if that were the case, soldering psram probably shouldn't make things worse since you can overclock the clock to 330MHz on 25q128 alone. Of course, I tested psram at 150MHz and it doesn't work. |
||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |
