DIY v2.5 Motherboard mod?

Hello all. I was re-reading the March update on Crowd Supply, and noticed the following piece about the new v2.5 revision of the Reform motherboard (Numbering added by me):

A revised v2.5 MNT Reform motherboard is also done, fixing some bugs with the initial design as well as switching parts out based on availability. The main improvements are:
1: Lowered headphone lowcut filter for improved bass response
2: USB-C replaces microUSB as the LPC flashing connector
3: Lowered power LED brightness
4: Fixed current leak through balancing circuit that could bypass missing cells
5: Fixed main buck converter enable behavior under low voltage condition
6: Fixed charger control circuit that previously required a factory resistor bodge

I began pondering the feasibility of applying some of these tweaks to my existing (v2.0?) to create a blend of both. (v2.25? v2.45?).
I don’t care that much about some of them. #2 isn’t important to me, it would be a very involved fix and wouldn’t achieve any notable benefit. Same with #6, I don’t mind the bodge at R53. It just reminds me that this was built by humans, too :^)

#1, #3, #4 and #5 all seem quite interesting to me.

I know #1 should be feasible. C147 was changed from 47µF to 220µF (with a helpful label to make it easy to spot :D). It’s in a bit of an awkward spot between the Ethernet jack and the headphone jack, but my TS100 is slim enough to work there.

I know #3 is also certainly feasible, just replacing R164 and R33 with higher value resistors should do. I’m not sure exactly what the change was, though. I tried comparing the schematics between the old and new versions, and both versions specified 1k resistors. Maybe the diodes were changed?

#4 might be feasible? I’m not sure how critical this fix is, but I see a 100k resistor (R86) added between Vref and Vtemp1 of the balancer, and R79 was maybe changed somehow? I don’t know what all the symbols mean, I’m very new to reading schematics!

#5, don’t know. It seems like Q13 and Q14 near U7 and U12, respectively, were both swapped for a resistor and diode pair? Are Q13 and Q14 transistors? I’m not sure. Maybe feasible. Again, don’t know how important this fix is.

Regardless, I most care about #1 and #3, and those seem the easiest to do.

So I’m hoping we could document some of those here to figure out how feasible applying these at home with decent soldering skills would be. I might be able to find a hot air station at the local hacklab too, it’s been 12 years since I did any significant SMD work, and these components are tiny!
I also don’t know how reasonable it is to acquire these tiny passive components in small quantities. Are there any nice assortments of these SMD parts available in varying sizes and specifications? It would be nice to stock up on lots of different ones at once, just to have them on hand. I’ve been meaning to put together a more permanent electronics workstation in my home office anyway.

Thanks for reading! I’m excited to learn more about these changes, and trying these out on my hardware.

1 Like

For #1 it was based off my original modification from Speakers too Quiet? Try This - #4 by Chartreuse where I replaced the two 47uF MLCC capacitors with first 100uF then 330uF electrolytic capactiors to upgrade the bass response. It’s a bit fiddly soldering those through-hole caps to the tiny 0603 pads, and you then need to make sure they’re solid and supported as to not rip those pads off afterwards.

220uF is a good compromise on size as with say 24ohm headphones (16-32ohm is common for consumer ones) that’d give a bass rolloff of 30Hz (-3dB point), much better than the stock 141Hz

You’ll want to change both C127 and C147 otherwise you’d have an imbalance between the two stereo channels.

A simpler version is what @mntmn did as a quick test which is to piggy back another 0603 47uF ceramic capacitor ontop of the existing one (Rolloff of 70Hz), though it’s still a little fiddly to do, but not as large. And they later added another on top to bring it up to around 141uF ish. (Rolloff of 47Hz)

Why does #6 not appeal to you? It is a little vague here, but I’m curious what exactly is the issue here that is being corrected?

Number 6 was more of a cosmetic issue, elimination of a bodge we had to do on each board.