gingernuts
Member
 Offline
Activity: 89
Merit: 10
|
 |
September 05, 2013, 10:02:50 PM |
|
I can't help wondering though, has anyone had success with fewer, larger caps? From what I can see, all the grounds are on the centre pad, all the pins on 3-sides are commoned internally to VDD, then split out to dozens of tiny caps, via'ed down to the ground plane. Does this have some inductance or ESR benefit, are the caps just cheaper, or is this just a design that works, so why mess with it?
More little caps in parallel have lower inductance than a few big ones. Maybe 15 caps is overkill, but it works, so why mess with it ? I could try mounting fewer, but the problem is that if it doesn't work properly, I'd have to hand solder the rest, which is a terrible job. Fair enough! Cheers. |
|
|
|
|
|
|
|
|
|
|
|
|
|
It is a common myth that Bitcoin is ruled by a majority of miners. This is not true. Bitcoin miners "vote" on the ordering of transactions, but that's all they do. They can't vote to change the network rules.
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
|
|
ssi
Member
Offline
Activity: 70
Merit: 10
|
|
September 05, 2013, 10:10:35 PM |
|
Thanks for the info... I've already got 100 of these boards being fabbed, but I'll take all this into consideration for the next iteration  Are you going to sell the things? Assuming they work, and assuming I can get the heat out of them, and assuming I can get chips, yes If you're interested in bare boards for DIY, lemme know in pm |
18xEDfc7y1Nzm2kmLvwYq56xwwEz4Fdh6
|
|
|
|
Gomeler
|
|
September 06, 2013, 05:45:33 AM |
|
This is my first attempt at a single-chip schematic. No joke, I was more or less copying the c-scape design for their initial test PCB. Electrically I think this is all correct? I'm going with the same decoupling capacitor values* that were mentioned in a post by intron.  I couldn't figure out how to add a ground pad for the 5GHash chip but in the PCB design phase I'm going to have a copper net underneath the chip with vias to the ground plane for heatsinking and ground purposes. I plan on using a 4-pin female molex connector to provide ground/VDD/IOVDD and a 8-pin header for the various I/O pins. After I layout this I'll have a second PCB with dc-dc converters to provide IOVDD and VDD. That same PCB will have the level-shifter to shift the spi signal voltages between the Raspberry Pi and 5GHash voltages. My goal is to dip my toes into this to see if I enjoy it. So far I do but I haven't had to debug anything yet  I'd like to chain 2 of these together and then build a 2 or 4 chip PCB using the lessons learned from this project. *- I found a couple very interesting PDFs on calculating the decoupling capacitance and the resulting inductance. Still educating myself on this so I decided to go with values that have been proven to work. |
|
|
|
|
intron
Sr. Member
Offline
Activity: 427
Merit: 251
- electronics design|embedded software|verilog -
|
|
September 06, 2013, 08:15:34 AM |
|
This is my first attempt at a single-chip schematic. No joke, I was more or less copying the c-scape design for their initial test PCB. Electrically I think this is all correct? I'm going with the same decoupling capacitor values* that were mentioned in a post by intron.
Looks fine. Toch? intron |
|
|
|
|
|
cscape
|
|
September 06, 2013, 08:19:44 AM |
|
Maybe add a solder jumper from INCLK to GND.
As far as calculating bypass caps, to do a proper job you need to know the layout of the board and the exact specifications of the capacitor. As they say: "ain't nobody got time for that!"
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
ssi
Member
Offline
Activity: 70
Merit: 10
|
|
September 06, 2013, 02:07:15 PM |
|
Maybe add a solder jumper from INCLK to GND.
As far as calculating bypass caps, to do a proper job you need to know the layout of the board and the exact specifications of the capacitor. As they say: "ain't nobody got time for that!"
Or, just overkill the hell out of it |
18xEDfc7y1Nzm2kmLvwYq56xwwEz4Fdh6
|
|
|
|
cscape
|
|
September 06, 2013, 02:11:25 PM |
|
Right. You can't have too much overkill.
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
|
LaserHorse
|
|
September 06, 2013, 02:29:25 PM |
|
since we're on the topic - I'm assume switching to 0603 size caps doesn't pose a problem, correct? After hand soldering 150 0402 caps, I'm ready to not ever do that again  |
|
|
|
|
cscape
|
|
September 06, 2013, 02:33:20 PM |
|
0603 size has a somewhat higher inductance, so it may not perform as well as 0402. Whether that's significant, I don't know.
Soldering 0402 has one advantage though: after you pick and place 10000 of them, suddenly 0603 seem as huge as bricks.
|
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
|
LaserHorse
|
|
September 06, 2013, 02:39:44 PM |
|
0603 size has a somewhat higher inductance, so it may not perform as well as 0402. Whether that's significant, I don't know.
Soldering 0402 has one advantage though: after you pick and place 10000 of them, suddenly 0603 seem as huge as bricks.
indeed - i can now do 1206 with my feet! thnx, will read up a bit on package characteristics. |
|
|
|
ssi
Member
Offline
Activity: 70
Merit: 10
|
|
September 06, 2013, 03:05:11 PM |
|
since we're on the topic - I'm assume switching to 0603 size caps doesn't pose a problem, correct? After hand soldering 150 0402 caps, I'm ready to not ever do that again I used all 0603, and they're fine. My supply planes are nice and quiet. |
18xEDfc7y1Nzm2kmLvwYq56xwwEz4Fdh6
|
|
|
|
Gomeler
|
|
September 06, 2013, 04:11:54 PM |
|
Glad to hear I didn't screw up completely. Going to spend tomorrow morning placing/routing the PCB with the goal of submitting it to OSHPark for a run next week. Will report back once I have the gerber files. This is my first attempt at a single-chip schematic. No joke, I was more or less copying the c-scape design for their initial test PCB. Electrically I think this is all correct? I'm going with the same decoupling capacitor values* that were mentioned in a post by intron.
Looks fine. Toch? intron What is Toch? Didn't see anything like that mentioned in the chip pinout, not sure if I'm missing something for the SPI connections? Maybe add a solder jumper from INCLK to GND.
As far as calculating bypass caps, to do a proper job you need to know the layout of the board and the exact specifications of the capacitor. As they say: "ain't nobody got time for that!"
Excellent point for the INCLK pin. I don't see myself driving the chip with an external clock source so I'll pull it to ground. since we're on the topic - I'm assume switching to 0603 size caps doesn't pose a problem, correct? After hand soldering 150 0402 caps, I'm ready to not ever do that again I used all 0603, and they're fine. My supply planes are nice and quiet. I didn't fancy placing those tiny 0402 capacitors. Since I'm just doing a cheap OSHPark run of 3 boards, I hopefully won't lose too much if I run into issues. |
|
|
|
|
|
cscape
|
|
September 06, 2013, 05:14:07 PM |
|
What is Toch? Didn't see anything like that mentioned in the chip pinout, not sure if I'm missing something for the SPI connections?
"Toch?" is Dutch for "Right?" |
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
|
Gomeler
|
|
September 06, 2013, 05:36:40 PM |
|
What is Toch? Didn't see anything like that mentioned in the chip pinout, not sure if I'm missing something for the SPI connections?
"Toch?" is Dutch for "Right?" Oh, thanks for the translation. I barely have a handle on English and it is my first language  |
|
|
|
|
|
Tyger
|
|
September 06, 2013, 07:38:43 PM |
|
sounds like someone from The Netherlands, not?
of Belg kan natuurlijk ook.
|
|
|
|
|
|
vs3
|
|
September 06, 2013, 08:35:10 PM |
|
Project NanoFury moving along too:  I can hardly wait to get my hands on the boards ( ignore the somewhat iffy looking parts - I haven't figured out the proper 3d models yet) By the way - speaking of filter capacitors: I figured 3 on each side should probably work just fine ... I guess I'll find out if it really does. |
|
|
|
|
bigbeninlondon
|
|
September 06, 2013, 09:03:35 PM |
|
Project NanoFury moving along too: I can hardly wait to get my hands on the boards ( ignore the somewhat iffy looking parts - I haven't figured out the proper 3d models yet) By the way - speaking of filter capacitors: I figured 3 on each side should probably work just fine ... I guess I'll find out if it really does. Do you have a thread following this development? I'd like to follow. |
|
|
|
|
|
vs3
|
|
September 06, 2013, 09:20:18 PM |
|
Project NanoFury moving along too: I can hardly wait to get my hands on the boards ( ignore the somewhat iffy looking parts - I haven't figured out the proper 3d models yet) By the way - speaking of filter capacitors: I figured 3 on each side should probably work just fine ... I guess I'll find out if it really does. Do you have a thread following this development? I'd like to follow. Not yet ... I guess I should do that |
|
|
|
gingernuts
Member
Offline
Activity: 89
Merit: 10
|
|
September 07, 2013, 08:58:02 AM |
|
So is there any advantage to mounting all the caps to the backside of the board so that a heatsink can be bolted down to the topside of those chips, or does the centre pad provide enough heat transfer through the board to make PCB back-side cooling better?
|
|
|
|
|
|
cscape
|
|
September 07, 2013, 10:16:23 AM |
|
So is there any advantage to mounting all the caps to the backside of the board so that a heatsink can be bolted down to the topside of those chips, or does the centre pad provide enough heat transfer through the board to make PCB back-side cooling better?
These chips are designed to dissipate heat through the ground pad and to the bottom of the board. In practice, cooling the board works well enough. |
Happy with your c-scape product ? Consider a tip: 16X2FWVRz6UzPWsu4WjKBMJatR7UvyKzcy
|
|
|
|
