Topic: Community Miner Design Discussion - page 16. (Read 34221 times)

sidehack

I have a design requset:
Allow the hash board input voltage to be from 10.5Vdc to 15.5 Vdc.

Here's the reason:
An input V range of 10.5V-15.5V would allow the boards to be driven by a battery bank which is in turn charged by solar.
And eliminates the need for a high current sepic converter

Here's a sample scenario that would scale:
3x 60 cell panels (250W-300W) in series (nominal 105Vdc input) feed to a Morningstar MPPT-60 (other brands exhibit similar tolerances) that charges the battery bank.
In this scenario, one can expect or should design for, a battery bank voltage to swing from 10.5V (completely discharged) to 15.5V (fully charged in winter and/or battery equalization charge levels).

Driving a miner directly from a DC source (12V battery bank) has at least one advantage, it escapes the losses in converting DC to AC (i.e. inverter inefficiencies).
Inverter efficiencies can be as low as 82% (sometimes worse with cheap China originated inverters).


Another request, regardless of input source, is that the hash boards implement a low voltage cutoff @ 10.5V.
This would prevent an over current condition when the bucks are feed a low supply voltage.
Which would assist in saving the hash board when a PSU screws up or is overloaded.

Maybe a simple fuse inbetween the PCI power input and the buck(s) supply rail???

Thanks in advance for the consideration.

Pup Tentacle

Basically we are too far down on the Asic food chain to get really solid info.

  A guy like me yeah I am honest   I have a bit  of a following, but money talks and I don't spend  100,000 usd   3 or 4 times a year on gear.
I may buy 30,000 usd worth of gear  in a year.  Which gets me a following since I sell it to the community at little or no markup. No one gets cheated by me. I have no preorder delays.  But I would still be called a lightweight to a company like bitfury. If I consistently bought and sold a few hundred thousand I would be more in the loop.


So I am always a little out of the loop and never quite have perfect info.

Yes, that i mean. With their demo video, they stopped the market sales; everybody is waiting for them. I don't think Bitmain is very happy with this videos, but Avalon is worse. The Spoondolies' guys are out of market with this anouncement. With theese chips, only Bitfury will control the entire markey (with permision of bw.com)...

Good ol' BE300, December 2014. Those were the chips that made me and Novak decide we should design our own miner. Too bad ASICMiner imploded right around then, because their chips tested out better numbers than pretty much anything until the BM1385 was released. Would have trashed the S5 and SP20.

I do not understand you.  They ran the chips on youtube videos . So the chips exist .

Now a while back Friedcat ran some really good chips but they never got to the public.

So are you asking is it sure bitfury chips will come to the public?

Simple answer maybe. 

Exact chance that the public gets chips  my guess is better then 80-85%.

It is for sure that bitfury HAVE theese chips? or they are winning time?? 

Getting back late to the discussion, so apologies if I state what has already been said.
I'll take the debit for the community ASIC discussion death.
I got sick and basically fell off the planet several months ago. Sucks to get old, BTW. Don't do it, if possible.

At a minimum, the cost to bring an ASIC to tape out is $1.5m
NotFuzzyWarm is spot on in his post.
From a business standpoint, for an entity to layout $1.5-$2.0 mil requires some surety of sales/profit.
If for no reason other than being able to continue operations (i.e. net revenue = $0).
On paper / in posts, it seems that BF has "stepped up to the plate" regarding chip availability.
Time will tell if/when chips actually make it into the community's hands.
And while I'm hopeful, I'm not "holding my breath" waiting for it to happen.

IMHO the kilo17/GS alliance is the communities best bet for affordable miners to empower the continued de-centralization of BTC.
For what little it's worth, we support this endeavor and look forward to it's success.

Also from a business standpoint, kilo17's financial participation is a gutsy move in an unsure industry.

I really am hoping hard on this.  The solar array will be ready in 10 days or so. I think we are going to run it at 6-7kwatts and wait for some bitfury gear(9-10kwatts) rather then load it up with avalon 6s and s-7's.

At least this is current plan for today.

well it means alot coming from you i know from seeing you around here you have been involved in darn near every reputable company for ascii's as a purchaser, so that makes me feel better, yay!
it's just FELT like bitfury only sells us stuff when they need money last year or 2 and rest of time they private miners ;\
i mean my one big point on that is that bit-x cloud mining is "unprofitable" for months now? thought they were bitfury backed.. you know what im saying? (not wanting to cut in community)
time will tell, but i believe if you do, and have my btc waiting

Right now I'm trusting kilo17. If he trusts Bitfury, or whoever is the gobetween he's using to get chips, then I reckon we'll have chips. Anyways he's moving forward with his own money so I can design around 'em, so he better be right.

I hope you're right

I have faith in Bitfury they will provide us with a solid amount  of chips at a decent price and earlier

then  expected.


How is that for a BOLD  remark?

I did not realize that sidehack did not get the chip info yet.
They will for sure provide a sample as soon they get more chips , at least to have more options for distribution later

arg, i personallly am getting discouraged in bitfury, not that it surprises me ;\
they seem to have the time for the big money and deal with bcn, but not sidehack
i really hope they honor their word but i fear no chips till april or later ;(
cmon bitfury, don't let us down, choose community for once!

Well I don't have any Bitfury chips to design a stick around yet...

i dont think hacks going to work on another stick design now untill this project is done but im sure sidehack will chime in on that

and sticks maybe?

S1 compatible? Oh boy.. I have 6 s1s in my closet, all gettin dusty. It'd be great to put them back in working order.. I just love the bare pcb look of the hashing units on s1s

I really hope this goes ahead, it would be really nice to have a small efficient home miner .

Well as has been noted fairly extensively so far, the primary hardware goal is to build something S1-compatible. So, about 15x25cm