I think you've got another winner there, Brian. I'm in for two boards -- and you don't even have your prototypes yet!!! HAH!!!
Only thing I might suggest (and you might already have done this) is to pick a box that it's intended to be in (CG1000?) and make sure the mounting holes match-up.
dirknerkle wrote:Only thing I might suggest (and you might already have done this) is to pick a box that it's intended to be in (CG1000?) and make sure the mounting holes match-up.
Yep - the bottom holes fit OK into the CG-1000 as before but am looking at the top holes - I'm hoping to at least get one of the top holes lined up so there will be three points. I'm also looking at revisiting removing +DC from pin 7 so you can run AC/DC SSRs.
Oh - - and the next set of boards will be tinned... :rolleyes:
Switchers, even the "simple switcher" types can be rather finiky. You need to use fat, low inductance traces or pours for the connections, and you need to minimize the switching loop areas. I would move the diode, inductor, and output cap closer to the regulator, and would rotate things to make routing cleaner. Also, the output capacitor is connected to the wrong side of the inductor. This layout is a bit cleaner, though it may be better to run the feedback trace from pin 4 to the output cap on the backside under the groundplane for some shielding.
/mike
n1ist wrote:Switchers, even the "simple switcher" types can be rather finiky. You need to use fat, low inductance traces or pours for the connections, and you need to minimize the switching loop areas. I would move the diode, inductor, and output cap closer to the regulator, and would rotate things to make routing cleaner. Also, the output capacitor is connected to the wrong side of the inductor. This layout is a bit cleaner, though it may be better to run the feedback trace from pin 4 to the output cap on the backside under the groundplane for some shielding.
/mike
Very nice - thanks for that Mike - I'm still massaging things around so I'll make those changes. I'm using this coil for the circuit - I overrated it a bit (1.9A) - maybe too much as it's fairly large. I have the parts in hand and have the space (increased board size to a full 4x7 from 4x6.55) so perhaps not a big deal.
Also - what filtering do you suggest for the +5v bypass input? Would the single 330uF cap be sufficient or should I have an alternate if going that route? I've recommended nothing but a good regulated 5v supply as the input when doing this. I'd like to keep with the same parts but if not, I could add secondary pads for a larger cap if necessary.
Re: Ren48LSD v3c
I'd be in for 2-4 PCB's and parts for at least 2 fully populated boards if/when the time comes for the next co-op. Now to try and get some of the LED strips...
Thanks,
-joni-
budude wrote:I've started laying out the next version of the Ren48LSD (v3c) with the following enhancements:Completely different voltage regulation circuit using the LM2575 switching regulator. The circuit comes right from the application sheet so it's pretty straightforward. I added an additional diode in the path to prevent damage if the power supply is installed backwards (idea stolen from RPM's DMX DCSSR16 circuit :rolleyes:). There will be no heat sink required and it will run up to 24vdc.
There will be a separate terminal block to bypass DC IN 1 direct to the 5v rail. Previously you had to solder in a wire across the original 7805 regulator pads - now you just have to install a jumper wire onto the block. I could have put a regular header but thought that might be unsafe (i.e. too easy to do...).
Aside from that I didn't change much - I cleaned up the silkscreen and added some nomenclature to make debugging easier.
Tne transistor pads were enlarged a tad to ensure they fit OK after they are tinned (suggestion from LabRat).
For those of you that have the Ren48LSD if there any suggestions I'll see if they can be added to the board. There will be another coop next month if there is enough interest. I'm still working on the final version but should be done soon to send out for a set of prototypes.
