Klondike - 16 chip ASIC Open Source Board - Preliminary

[BkkCoins]

[…]
To aid with inspection of test and assembled boards I'm going to order one of these USB Microscopes. It looks pretty nice for checking PCB traces and IC pins, and my eyesight is not what it was (as I realized while looking at some boards that I got yesterday). For $60 it seems like it'll be worthwhile.

http://www.youtube.com/watch?feature=player_embedded&v=2te13V_YsPU

Anyone ever use one of these?

I have used a very similar model.  They're certainly adequate for spotting solder bridges on SMT devices.  
I found the LEDs a bit harsh and the image quality a bit lacking, though this is likely unimportant for your usage.

EDIT: Also this model looks quite lovely & capable:
http://www.ebay.com/itm/Mini-2MP-USB-Digital-Microscope-Circuit-board-Inspection-Camera-Aluminium-Stand-/190847055453?pt=LH_DefaultDomain_0&hash=item2c6f5ef65d
likely same as this one: http://youtu.be/YDCicFZEINo?t=15s

That one looks awesome. I need it.

BKK, if you agree that this one will work for your purposes, send me your address, and I'll purchase one and have it shipped to you. 

Nice! That one is really beautiful and I like the stand as well. That seems essential for steadiness.
Thank You and I'm sending you a PM.

[TomKeddie]

Nice! That one is really beautiful and I like the stand as well. That seems essential for steadiness.
Thank You and I'm sending you a PM.

I use one of these visors for all my work now, I have a stereo microscope but I really just prefer these.

http://www.amazon.com/dp/B0015IN8J6/ref=biss_dp_t_asn

[SebastianJu]

Anyone know how the chips are supplied in reel or tray?
In avalon specs do not mention this  

I'm sure it won't be in a reel; most likely a tray,

The samples might be in a tube.

I found these pictures from asicminer asic's. https://bitcointalk.org/index.php?topic=99497.msg2268332#msg2268332 Bitfountain uses TMSC foundry too. And friedcat told me "Our chips use 6mmx6mm QFN40 package." while avalons are 7x7mm.

So i think its possible that all chips come in tubes. I mean the tubes could easily be put into a packet and shipped. And i think the reason why the sample chips per batch are exactly 30 are that the tubes contain 30 chips... at least AM's tubes contained 30 too.

I have to know too how avalon ships because i have to buy the needed packaging for my groupbuy...

[BkkCoins]

Nice! That one is really beautiful and I like the stand as well. That seems essential for steadiness.
Thank You and I'm sending you a PM.

I use one of these visors for all my work now, I have a stereo microscope but I really just prefer these.

http://www.amazon.com/dp/B0015IN8J6/ref=biss_dp_t_asn

That's also a good idea and I think I've seen them in the Bkk street markets so I'll grab one next time I'm there. They have this large city block sized place called Klong Thom here that is chock full of industrial/consumer stuff, though it's a really congested experience, especially on Sundays when it seems to double as a "swap meet" around it's outside.

[BkkCoins]

So i think its possible that all chips come in tubes. I mean the tubes could easily be put into a packet and shipped. And i think the reason why the sample chips per batch are exactly 30 are that the tubes contain 30 chips... at least AM's tubes contained 30 too.

I have to know too how avalon ships because i have to buy the needed packaging for my groupbuy...

I think that's quite likely as well. (And probably annoys board assemblers)

I hope zefir and others will keep packaging on re-shipment to a minimum eg. tape chips to card board and put in a padded envelope. I once received a couple sample chips from Xilinx on a FULL size tray, in a large box, shipped by FedEx. I was just glad they covered that and not me.

I've received Core2Duo CPU chips squished between foam squares in a padded envelope a couple times without any problem so I'm sure QFNs don't need excessive packaging.

[SebastianJu]

So i think its possible that all chips come in tubes. I mean the tubes could easily be put into a packet and shipped. And i think the reason why the sample chips per batch are exactly 30 are that the tubes contain 30 chips... at least AM's tubes contained 30 too.

I have to know too how avalon ships because i have to buy the needed packaging for my groupbuy...

I think that's quite likely as well. (And probably annoys board assemblers)

I hope zefir and others will keep packaging on re-shipment to a minimum eg. tape chips to card board and put in a padded envelope. I once received a couple sample chips from Xilinx on a FULL size tray, in a large box, shipped by FedEx. I was just glad they covered that and not me.

I've received Core2Duo CPU chips squished between foam squares in a padded envelope a couple times without any problem so I'm sure QFNs don't need excessive packaging.

The tubes and reels have ESD-Protection already. So i thought to put the tubes (the ends closed with a little bit ESD-bag and ESD-Tape) into a normal bag, put silica gel to it and ship it as packet in a air pad bag. The reel-enclosure is esd-protected too. And all chips that go to an assembler go with one shipment together. I think this way the chips are protected against ESD, humidity and mechanical pressure.
But i think i will keep them in the tubes. Its not good for automatic assembly but i cant put it on a reel too...

[fasmax]

@BkkCoins
It looks like the PIC16LF1459-I/SS in the TSSOP is in short supply do you have a source for this part?
Is there similar PIC chip that might work in its place?
Thanks!

[sensei]

So...
Just ordered first run of 10 test boards for K16.

And SDS7102 Oscilloscope ordered today too, direct from China. They tell me it'll ship by EMS so it should only be about a week to get here if all goes well.

So the way things are looking now I'll have enough to test PIC firmware in a few days. Should have parts needed for testing K1 boards when they arrive in a week, and maybe scope will be here for that. That gives me a week to play with that stuff and get kinks worked out before K16 boards come in, and then I should have everything here to get that scoped and functional.

And if the planets align that will all be just in time for the samples to start arriving, and the real fireworks happen.

To aid with inspection of test and assembled boards I'm going to order one of these USB Microscopes. It looks pretty nice for checking PCB traces and IC pins, and my eyesight is not what it was (as I realized while looking at some boards that I got yesterday). For $60 it seems like it'll be worthwhile.

http://www.youtube.com/watch?feature=player_embedded&v=2te13V_YsPU

Anyone ever use one of these?

I was looking at getting the same one. I'm glad I watched this review. I was a bit hesitant, but not anymore.

[allten]

So...
Just ordered first run of 10 test boards for K16.

And SDS7102 Oscilloscope ordered today too, direct from China. They tell me it'll ship by EMS so it should only be about a week to get here if all goes well.

So the way things are looking now I'll have enough to test PIC firmware in a few days. Should have parts needed for testing K1 boards when they arrive in a week, and maybe scope will be here for that. That gives me a week to play with that stuff and get kinks worked out before K16 boards come in, and then I should have everything here to get that scoped and functional.

And if the planets align that will all be just in time for the samples to start arriving, and the real fireworks happen.

To aid with inspection of test and assembled boards I'm going to order one of these USB Microscopes. It looks pretty nice for checking PCB traces and IC pins, and my eyesight is not what it was (as I realized while looking at some boards that I got yesterday). For $60 it seems like it'll be worthwhile.

http://www.youtube.com/watch?feature=player_embedded&v=2te13V_YsPU

Anyone ever use one of these?

BkkCoins,
     Are you sure the firmware can handle everything ok? It only has 1Kbytes of SRAM.
I'm about to order boards for the Quarter Stick DIY and I wanted to make sure.

thanks.

[BkkCoins]

BkkCoins,
     Are you sure the firmware can handle everything ok? It only has 1Kbytes of SRAM.
I'm about to order boards for the Quarter Stick DIY and I wanted to make sure.

thanks.

I believe so. Work so far indicates that it will be very close with a work queue of 4 items (192 bytes), and a 108 byte buffer for pushing work. So if it gets too tight for that I may have to drop to only 2 work items queued. Compiled with the USB stack it's using about half the RAM, without my code and work queue, (and 727 bytes with my code but no push buffer yet), but I also think that could be optimized and reduced a some if really need be (the stack seems a little bloated in C instead of assembly. I can't fathom why they didn't do it in assembly for something that could be included in every PIC - though maybe they expect their users to optimize their own versions). By slowing down the push-work a bit with a buffer switch mid-way, I could cut 48 bytes and keep the queue at 4 but I haven't gone over everything looking to optimize RAM.

Also, I haven't tried the Pro XC-8 yet. Apparently it cuts down a lot of code space, though whether it can get much better at RAM use is questionable. I'll have a better estimate later this afternoon after I integrate in the "push work" assembly code. I'd code everything in assembler before giving up.

[dvolosk]

Posting in this thread to say that I love you BkkCoins.

Any Texas based users here planning on having some boards locally fabricated?

[marto74]

I think that's quite likely as well. (And probably annoys board assemblers)

Just a few words off
I beleive that Pick and Place machies have specialized feeders for tubes.
I made a few custom ones a few years back.
These were basedon vibration.

[allten]

BkkCoins,
     Are you sure the firmware can handle everything ok? It only has 1Kbytes of SRAM.
I'm about to order boards for the Quarter Stick DIY and I wanted to make sure.

thanks.

I believe so. Work so far indicates that it will be very close with a work queue of 4 items (192 bytes), and a 108 byte buffer for pushing work. So if it gets too tight for that I may have to drop to only 2 work items queued. Compiled with the USB stack it's using about half the RAM, without my code and work queue, (and 727 bytes with my code but no push buffer yet), but I also think that could be optimized and reduced a some if really need be (the stack seems a little bloated in C instead of assembly. I can't fathom why they didn't do it in assembly for something that could be included in every PIC - though maybe they expect their users to optimize their own versions). By slowing down the push-work a bit with a buffer switch mid-way, I could cut 48 bytes and keep the queue at 4 but I haven't gone over everything looking to optimize RAM.

Also, I haven't tried the Pro XC-8 yet. Apparently it cuts down a lot of code space, though whether it can get much better at RAM use is questionable. I'll have a better estimate later this afternoon after I integrate in the "push work" assembly code. I'd code everything in assembler before giving up.

You're a trooper.  Thanks for letting me know. I think I will get some boards
and sleep better at night knowing they have a better chance of working.

After that, I'm hitting the code.
FYI, I have access to a Pro XC-8 compiler; if you need me to compile anything
let me know.

What are your plans on sharing the source development? If you plan to keep it internal,
that is fine. Would just like to know to avoid reinventing the wheel.

[BkkCoins]

You're a trooper.  Thanks for letting me know. I think I will get some boards
and sleep better at night knowing they have a better chance of working.

After that, I'm hitting the code.
FYI, I have access to a Pro XC-8 compiler; if you need me to compile anything
let me know.

What are your plans on sharing the source development? If you plan to keep it internal,
that is fine. Would just like to know to avoid reinventing the wheel.

I just tried with 4x work queue, pre_calc 6 words, 8 nonce range words and 2x result queue in RAM (plus other core stuff like USB buffers, and status and cfg registers). I have more code to do but I think this is all the main data chunks required other than the odd temporary byte here or there. RAM usage is currently 805 bytes. I'm fairly comfortable that we're good for RAM.

I may take you up on that XC-8 Pro offer later. It may well be crucial to fitting the code in 8K. I can enable the trial 60 days but then why do that until absolutely required (or I'm curious if you can do that in a VM and then wipe it after expired?).

I'm going to release the firmware and driver code once it's working. I was thinking of all that stuff to protect it but to be honest that's just is a hassle. I'm going to just trust that people will choose to support the vendors that kick back a small fee. Or even better maybe they'll mine a few hours for me by their own choice. I'll put the mine-to address out there when it's time. I'm getting a lot of help/support from forum members here and I think it's better to respect that by not having locked code at all.

[mxz280]

Each K16 have a power connector, power plug or suggestions 2.5/5.5mm Type D Molex power connector.

[BkkCoins]

Each K16 have a power connector, power plug or suggestions 2.5/5.5mm Type D Molex power connector.

Each K16 has a PCI Express power connector. That's the 6 pin standard one seen on most PCI Express cards, and commonly available on ATX power supplies. Molex adapters are cheaply available, as are PCIe splitters, but be sure to use ones that have thick enough wire to handle 4A without getting hot.

[mxz280]

Each K16 have a power connector, power plug or suggestions 2.5/5.5mm Type D Molex power connector.

Each K16 has a PCI Express power connector. That's the 6 pin standard one seen on most PCI Express cards, and commonly available on ATX power supplies. Molex adapters are cheaply available, as are PCIe splitters, but be sure to use ones that have thick enough wire to handle 4A without getting hot.

ATX power only two PCI Express power connectors, how can connect 16 K16?

[evilscoop]

time to get to work on a power distribution board me thinks.....
enterpoint did one for the cainsmore, might be good to get something similar made up

would something similar to this work :-
http://www.enterpoint.co.uk/cairnsmore/CAIRNSMORE1_POWER_DISTRIBUTION_CAD2_674X388.JPG

[BkkCoins]

Each K16 have a power connector, power plug or suggestions 2.5/5.5mm Type D Molex power connector.

Each K16 has a PCI Express power connector. That's the 6 pin standard one seen on most PCI Express cards, and commonly available on ATX power supplies. Molex adapters are cheaply available, as are PCIe splitters, but be sure to use ones that have thick enough wire to handle 4A without getting hot.

ATX power only two PCI Express power connectors, how can connect 16 K16?

Do I really need to answer that?

sɹǝʇʇıןds ǝsn

[Biomech]

Each K16 have a power connector, power plug or suggestions 2.5/5.5mm Type D Molex power connector.

Each K16 has a PCI Express power connector. That's the 6 pin standard one seen on most PCI Express cards, and commonly available on ATX power supplies. Molex adapters are cheaply available, as are PCIe splitters, but be sure to use ones that have thick enough wire to handle 4A without getting hot.

ATX power only two PCI Express power connectors, how can connect 16 K16?

I'm new to this myself, but I think you can buy a molex to PCIe adapter.