Klondike - 16 chip ASIC Open Source Board - Preliminary

[Igor_Rast]

I really hope the next project is on Eagle. I really don't like Kicad much.

Maybe we could take up a collection to buy an Eagle license for Bkkcoins

A little more then $1,600 bucks for the full featured license...im sure the community can scrape together 16BTC.

I'm not a big fan of Kicad but the main reason I used it was so that anyone could edit and work with the files without having to buy a license of any sort.  So I'm not really looking for a commercial version of Eagle. You have to remember that where I live I could have easily picked up Altium or Eagle for only a few $. Using Kicad was more a philosophical issue. Despite the troubles I had on the way I'm actually kind of impressed that it actually does work. I would change a bunch of things if I had the time and energy to contribute or fork the project - but I don't for now.

+1 for that BKK , in the end the Klondike design is opensource , you might as well use an opensource pcb router program to build it .

kicad took me allot of headace to get it working like i want , but once you get it right there is no going back to licensed Eagle , Same goed to windows after having to install ubuntu about 6 times over to get my screen working :p

Crowd Knowledge works best when not restricted , So opensource is the way of the future

 

[sensei]

With regard to the additional power needed for overclocking; how about scrapping the pads for the Phoenix power connector and placing the 9A regulator there. There could be some 0 ohm jumpers to take it in/out of the circuit. Then the single board could be used giving the user the option to go either way.

Just a thought.

[SteveB]

What is the input voltage range of the k16 board? Does it have to be exactly 12v or is there a range?
The reason I ask is that I have a small solar system (430 watts) and I wonder if it's possible to connect a k16 board directly to my 12v battery bank.
The voltage varies between 11 - 15.5 volts depending on battery charge and load level.

[marto74]

For sure You have to supply stabilized Voltage input in the range of standard ATX PSU

[fasmax]

I really hope the next project is on Eagle. I really don't like Kicad much.

As an assembler/fabricator, I will be ordering almost 200k resistors and capacitors within the next. week. I have spent many hours going through the BOM looking for the best value from countless suppliers and looking at alternate parts as some parts have 13 week lead times. There is one part that for a run of 500 K16's, I need almost 100,000 of them.

My son and I will be building 50 of the K16's for ourselves and I will run most of them separately on USB hubs. A few we will klego together just to prove out the capability.

Hold off on R47, C274 as they are changing. R47 may be dropped or moved.
Make sure you check current parts list as R45, R46 just changed.

The 0.47 uF cap is a nuisance as it's high qty tends to push you out of the stock-on-hand category, but there are quite a few alternates that could be used and some have typically had quite high stock levels.

I think that's all I have for the moment.

The reference design uses 470 ohm for R45 and R46. Wondering once you place all 16 Avalon chips if you might need a lower resistance to ensure a decent rising edge on the report signals. Next time you are looking at signals can you measure the rising edge of one of the report signals? I am just curious how many nano seconds its takes for the report signals to go from low to high with one ASIC installed and the 1K ohm pull up.
Thanks!

[BkkCoins]

What is the input voltage range of the k16 board? Does it have to be exactly 12v or is there a range?
The reason I ask is that I have a small solar system (430 watts) and I wonder if it's possible to connect a k16 board directly to my 12v battery bank.
The voltage varies between 11 - 15.5 volts depending on battery charge and load level.

The regulators will work up to 21V. Component selection may have issues with input below 12V. I don't think the input voltage would be a problem but stability or noise could be. You could test it and see but that would be entirely at your own risk. In theory it could work and shouldn't damaged by the voltage levels indicated - you may want to clamp voltage to ensure it cannot go higher.

[Bicknellski]

******
My current opinion / standing on the over-clocking power issues is that I'll get the current version working fully and then do a version 2 board that uses the IR3847 buck reg chip instead. The circuit is almost identical and just has some position/orientation changes. The sourcing may be more difficult but it's still the better choice for boosting the on-board power without too much redesign. Most of the parts are value changes to similar parts. I think the main issue will be the Cyntec 0.215 uH inductor (PCDC1008-R215EMO) and for those looking at sourcing and supply I'd start digging into how to get test samples and place larger orders of that part. It doesn't seem to be commonly available at Mouser/Digikey et al.

Just some forward looking comments at this point.

On the K1 it's possible to replace the 6502A with the 6503A and use a powered hub to get up to 3A of juice for the ASIC over-clocking. It's almost the same price anyway but would require a small bump in inductor choice.

http://www.canics.com/searchresult.php?q=PCDC1008-R215EMO

http://www.alibaba.com/product-gs/773050720/_Inductor_PCDC1008_R215ELO.html

Let me know if you want 10 I can order them.


Supply Ability:    800000 Piece/Pieces per Day

[Bicknellski]

Would I be correct in assuming the tiny I2C boost jumper is something like the one pictured below?

The K64 is just 4x K16s on one board so the two can interconnect as if the same. For now I'm considering the limit on chaining the boards to be 8 K16 boards. I think that's a dependable number. However, it could easily be more as some people report having 25x I2C LED boards on an bus. I'll do testing when some are built. Regardless, at some point I'll put together a tiny I2C boost jumper that slips over the pins joining boards and expands this number. Also, initially I'll stick with 8-bit addressing and on I2C I think this means just a bit under 128 boards could be chained in theory, off a single USB connection. I doubt anyone will do that, and it probably makes more sense to run about 16 boards off a USB, and then use a USB hub to expand further, seeing as how they are dirt cheap and let you use the higher bandwidth of USB vs. I2C. - BKKCoins https://bitcointalk.org/index.php?topic=190731.msg2107060#msg2107060

https://www.samtec.com/my-samtec/pricing.aspx?partnumber=MNT-108-BK-G




http://www.samtec.com/documents/webfiles/cpdf/Mnt-mkt.pdf

REVISION:

Would the connections need to be like this?



Premium Female/Female Jumper Wires - 40 x 6"



Handy for making wire harnesses or jumpering between headers on PCB's. These premium jumper wires are a little over 6" (150mm) long and come in a 'strip' of 40 (4 pieces of each of ten colors). They have 0.1" sockets on either end and fit cleanly next to each other on standard-pitch 0.1" (2.54mm) header. We recentlty updated these so they are in a 'ribbon strip' instead of individual wires. You can always pull the ribbon wires off to make individual jumpers, or keep them together to make neatly organized wire harnesses.

We suggest picking up some extra-long break-away header pins as well, you can stick those into the sockets to turn the cables into female-male or male-female jumpers!

[BkkCoins]

******
My current opinion / standing on the over-clocking power issues is that I'll get the current version working fully and then do a version 2 board that uses the IR3847 buck reg chip instead. The circuit is almost identical and just has some position/orientation changes. The sourcing may be more difficult but it's still the better choice for boosting the on-board power without too much redesign. Most of the parts are value changes to similar parts. I think the main issue will be the Cyntec 0.215 uH inductor (PCDC1008-R215EMO) and for those looking at sourcing and supply I'd start digging into how to get test samples and place larger orders of that part. It doesn't seem to be commonly available at Mouser/Digikey et al.

Just some forward looking comments at this point.

On the K1 it's possible to replace the 6502A with the 6503A and use a powered hub to get up to 3A of juice for the ASIC over-clocking. It's almost the same price anyway but would require a small bump in inductor choice.

http://www.canics.com/searchresult.php?q=PCDC1008-R215EMO

http://www.alibaba.com/product-gs/773050720/_Inductor_PCDC1008_R215ELO.html

Let me know if you want 10 I can order them.

Your a magician at digging up stuff. I'd love 10 if you can arrange it. That would save me time and be enough for testing. Maybe by the time they get here I'll be able ready to test a v2 board. Don't pay extra for rush fedex as this is still some time off and I'd need new boards first. Thx.

[BkkCoins]

Would I be correct in assuming the tiny I2C boost jumper is something like the one pictured below?

Close but not the same. You need to make connections across boards not just jumper pins, unless I'm reading that wrong. I'm using the parts below for testing, though other mfr. no doubt make the same. You want a 10 pin 0.1" spacing female header. I used SMD (but thru-hole will work) as I'll mount it on a tiny pcb that makes the connections. The middle 2 pins are n/c between boards.

Found and supplied to me by forum member mrm0. Thank You.

http://www.maritex.com.pl/en/shop/products/ggid/9327

CVILUX PLTD40S-CV  40 pin dual row SMD pin header
0.3256 EUR x 11 pcs/pkg
(I would cut these down to 2x2pin x 10 pcs each)
(I'm sure Greencom is same but they didn't have stock)

Greencom PLTS30R-LF  30pin horz single row SMD pin header
0.4428 EUR x 8pcs/pkg
(also would cut to 3 pin x 10pcs each)

Greencom PBTD10S 10pin female SMD header
0.2554 EUR x 10pcs/pkg
(these are for joining the pin headers, with one row gap between boards)

[loshia]

BKK,

Would you be so kind to update git and include latest version of cgminer driver?

I am curious to look at it.

Also do you think that hash precalc shall be done buy the PIC or cgminer? The prsumption is that there are much more resources in mining host compared to PIC

10X

PS: I am waiting for daily summary update impatiently

Best

[BkkCoins]

BKK,

Would you be so kind to update git and include latest version of cgminer driver?

I am curious to look at it.

Also do you think that hash precalc shall be done buy the PIC or cgminer? The prsumption is that there are much more resources in mining host compared to PIC

10X

PS: I am waiting for daily summary update impatiently

Best

The precalc is already done in the PIC and it works without using much resources.

I've been working on the cgminer driver. I had to revamp the PIC USB code to support LibUSB generic mode instead of CDC serial. That works now. The cgminer driver now detects the klondike and inits and adds it as a "cgpu". But after that it seg faults. So I'm currently tracking that down. There is no docs I'm aware of other than reading the code and it's complicated and convoluted. So I don't know off hand where and why it faults after that. I expect I'll have something before too long. It took me a long time to figure the usbutils code and why it wouldn't accept the PIC descriptors.

The rest of the cgminer driver is written but not yet tested. I'll wait until it works (even if not perfectly) before posting it to github. Anyway, it is coming along.

I've also recoded ktest to use PyUSB for libusb mode support.

The last line logged before seg fault is:

Code:

DBG: sending http://pool.50btc.com:8332 get RPC call: {"method": "getwork", "params": [], "id":0}

Though I have no way to know if that's related as there are a few threads running here. Just stopped to eat some dinner before tracking down this seg fault.

Note that when I switched to libusb communications the code and data size on the PIC dropped significantly. So I guess CDC Serial was wasting valuable resources. My last build was 4904 code and 631 data - which is way, way down.

[Chefnet]

ask ckolivas I think he will help you fast and he is the best as cgminer developer if you can grant him remote access.

[ryepdx]

ask ckolivas I think he will help you fast and he is the best as cgminer developer if you can grant him remote access.

+1. Delegate what you can. Your time is valuable!

[Bicknellski]

This is where a collection or BOUNTY from us sitting on the sidelines can be used to help.

+1 and delegation / remote help. If he wants a few sheckles to help let us know we will kick in some BTC for that.

Again many of us here will also be willing to put some burn in time directly to him for much needed development of the CGminer that works with your hardware BKKCoins.

[prodigits7]

I can vouch for ckovilas, he's the best at cgminer coding:

[gateway]

Also you can get good support for bfgminer on irc freenet #eligius , lukes been pretty good at helping out as well

[dentldir]

I'll pitch in as well to get ckovilas to help if he's interested.

[Bicknellski]

BKKCoins should we pass the plate and get some paid help from the king of CGminer ckovilas?

[BkkCoins]

BKKCoins should we pass the plate and get some paid help from the king of CGminer ckovilas?

Not really necessary - not everything happens instantly. Some things take time even when you don't have it.

Anyway - Good News - I have cgminer submitting shares now. It's still a bit rough as I figure out how the hash counting is supposed to work. It seems to spit out zillions of thread-hash count values when debugging and I had to disable that so I could see anything. Probably still something wrong there.

I'm finding libusb less stable than using serial mode. It's probably just that it's new code and I didn't spend much time with it yet. I'm seeing more errors and disconnects now. I'm also finding that after a while it seems the nonces seem to die out and I suspect it's got into some zombie state. Re-powering restores a trickle of shares again.

So I still have things to track down and fix, but the overall basics are there. It gets work, sends it to the PIC queue, ASIC hashes them, and results are sent back and accepted by 50btc pool.

Yay! To quote cgminer.

Code:

[2013-06-29 16:46:18] Klondike (5:11) reply [=:3d00001ea379be]
 [2013-06-29 16:46:18] Klondike FOUND NONCE (xbe79a31e)
 [2013-06-29 16:46:18] Klondike SUBMIT NONCE 1 (xbe79a25e)
 [2013-06-29 16:46:18]  Proof: 0000000075f0e30948e021c4fd561301dad5729c0ca0c491b1fb458dc0d1ee8f
Target: 00000000ffffffffffffffffffffffffffffffffffffffffffffffffffffffff
TrgVal? YES (hash <= target)
 [2013-06-29 16:46:18] Pushing submit work to work thread
 [2013-06-29 16:46:18] Creating extra submit work thread
 [2013-06-29 16:46:18] DBG: sending http://pool.50btc.com:8332 submit RPC call: {"method": "getwork", "p
arams": [ "000000028c3cf41eaee053c948489cf275b6dacbff0c74fc026552800000001c00000000b46c41b0e6a1acfe3e042
a44b07452e918efb4aab5ff0a0e767d95e4e5b6686d51ceac731a00de155ea279be0000008000000000000000000000000000000
00000000000000000000000000000000000000000000000000080020000" ], "id":1}
 [2013-06-29 16:46:19] (5s):251.5M (avg):266.5Mh/s | A:16  R:0  HW:6  U:0.7/m  WU:1.0/m
 [2013-06-29 16:46:19] PROOF OF WORK RESULT: true (yay!!!)
 [2013-06-29 16:46:19] Accepted 75f0e309 Diff 2/1 KLN 5 

Note that hash rate is wrong - I fudged the scanhash count with a const value for now.

I won't lie - It's been a hard 24 hours getting this to work.