Aside from the decreasing efficiency which for ATX supplies actually largely is from the fans running near/at full speed and taking more power to do it, with ANYTHING dealing with power be it plug/circuit ratings and including PSU ratings always follow the 80% rule. For the feeders it keeps plugs/cords/wiring behind walls at safe temps when under a constant near-max load - which miners present - for the PSU's you need that margin to handle the random (hopefully) short-term events all powerlines have.

For example of pushing ratings just talk to Bitmain and their experience with early s2 PSU's.... But as long as you have spares to cover ya while EVGA ships a replacement then your call. Und ja, as has been often said, any PSU is much happier on a 208-240v line because of the lower currents on the AC-line side of the power conversion.

If the resistor is put in the +12v line feeding the fan, sure it will work as a brute-force way to do it. Question is it a 3 or 4 wire fan? With either 3 or 4 wire it will be reporting it's speed, 4-wire gives it built-in PWM control.

With 4-wire most often no signal = fan ON (for safety and convenient way to let fan go full speed if wire 4 is open) and +5v = Fan min speed. Play with the signal duty cycle to set the speed between it's min and max speed.. Dunna know if the PSU's fan circuit does anything re: monitoring the speed except for maybe checking for a stalled fan.

My guess is inexperienced and locked into a 'well it worked before' mentality. Just gloming together more VR modules in parallel to give needed very high current and low voltage is just a small part of the battle.

Getting the desired power from the VR's cleanly to the chips is a whole 'nother ballgame and requires looking at the PDN like a radio circuit than a simple DC bus. Probing the power planes, the spiking rise/fall times and ringing you see on them even just an inch away from a different spot will be very different. To me at least, that is the biggest advantage to the string topology - currents are far lower making power plane considerations much easier to deal with.

Um, 'Not in the hot zone'?
The heatsinks being in contact with the chips sorta makes that non-sequitur as that is where majority of the heat is produced...

Agreed. The tickbox of required skillsets is extensive even if using an 'off the shelf' ASIC vs rolling your own.

Having been involved in forensics on the ill-fated Bitmine/AMT A1 boards it was clear whoever did the board design knew nothing about high current PDN's (Power Distribution Networks) much less thermal design. BFL's Monarchs are another example of failing on just those 2 points. Both Bitmine.ch and BFL apparently had no one in-house or consulting with any idea of just how different those power-hungry chips were from what they had dealt with before and what it meant to PDN and thermal considerations or if someone did raise flags they were firmly ignored.

Then yes there is the coms to work with. CGminer and using SPI of course works and is very flexible to modify/work with but... The underlying coms to/from the ASIC's leaves much to be desired. I know that chip-level IP is available to make each ASIC an address on a true local machine network so date rates to/from ASIC's and controller can now hit T1000 speeds if desired with the router/controller buffering what the mining server can handle/will allow. I'd think that ad-hoc addressing when needed vs fixed interval polling would be a far better way to handle the data today's chips can crunch.

I agree with Rabin - I'd reattach the heatsink myself using thermal epoxy as you said in the OP.

Where gravity is an issue, NEVER NEVER USE ONLY THERMAL TAPE!!! It is NOT made for structurally holding a heatsink on and if the sinks are vertical gravity/heat WILL cause them to start sliding and eventually fall off.

I suppose the biggest hurdle is getting viable ASIC's to use. That means either striking a deal with proven chip makers or going full ground-up custom.

As Bitmain proved with their revamped custom 28nm chips, using pre-packaged chip IP blocks that are cut and pasted into a mask then linked together to do the functions ala' Innosilicon and Bitmain's earlier chips is not the way to go. Most off the shelf IP blocks are to some extent fairly general purpose within their type of function and not specifically optimized just to crunch SHA.The IP blocks in question were developed by the various chip design houses like Inno and chip foundries like TSMC, GloFo, et al  to be reusable in different applications with 'acceptable' results.

That said, get licensed modification access to the IP in question so one can trim the fat and optimize the inter-block connections right at the block layout level and the results can be impressive..

It also calls for equally impressively skilled chip designers and deep pocket$$$.

Werks for me. Good to know there may be some use for the things.

Just to be clear, are you still looking for S1/S2 boards to use? And how about S3 boards? I have 2 s2's, 9 S1's complete and all retired along with over a dozen s3's I'd like to retire...
Trade-in value for them?

Define 'half the load".

Wattage rating for power flowing through a transformer is the same regardless of voltage.
Now the voltage or current ratings - THAT is what changes. Watts = Volts x Amps. When providing the same wattage, double the voltage and the current side is cut by 1/2 or conversely cut voltage by 1/2 and available current doubles.,

Interesting. My batch-1 s7 came with factory setting of 600mHz and is humming along perfectly at over 4.86 as reported by my pool. Guess I'll just have to goose the coming batch-2 one to the same and see how it does.

So happy so far with my Batch-1 s7 so I ordered a B2 one today. Wonder how long this one will take?  
Also - why are the batch3's a tad more expensive than the batch-2's?

 You don't draw double the power but could if you plug a 110v device into a 220v socket. Power is just a measure of work being done, and formula is Watts= E*I.

Motor windings, heaters, etc, are setup to work with a given voltage and for the same function (amount of work) will pull the same amount of power. Current draw will be different but total power drawn is the same. Watts is Watts.

As for why North America uses 110V, all I have been able to come up with is that when using 220v distribution it takes smaller gauge wires to support a given load and so it was cheaper. That or wire insulation methods had improved by the time the rest of the world started electrifying. Either way, 220v is far better.

1 week after the email saying it had been shipped my s7 arrived today.

Using an IBM 2kw psu and GeckoScience breakout/PCIe leads it is running perfect at between 4.86 and 5.1 Th/S per the GUI and averaging 4.8TH/s per my pool. Excellent.

Looks I'm going to need more breakouts and PCIe leads for when I get a couple more s7's... 

Well, our UPS deliveries for today have come and gone - no miner despite the email I got on Monday saying it has shipped  
Over 1 week in-transit??? Er, the other couple dozen miners I've gotten from the S1's on up which were shipped via UPS arrived in 3 days.

What did they do - send it on a cargo ship???

Faboo!! Grabbed. 

Ditto^^ Never EVER provide your SSN (for those not in the USA, Social Security Number) to a caller over the phone! Always find out who they are (in this case DHL), look up the local phone number for them and contact them yourself to be certain it is legit.

Same here. The email arrived 7 hrs ago and says it has shipped, no tracking info and also based on 9-28 ship date. Most odd... 

Also, when using a LAN straight from machine to machine you MUST use a crossover cable. not a 'straight-thru cable.. The LAN ports to not autoconfig what is the Tx and Rx lines. A hub has same problem, using any reasonably modern LAN switch between the miner and computer takes care of the issue as would a crossover adapter..

Novec is strictly from 3M and it's distributors. Thought about it myself but damn it is expensive... If BTC was still at last years prices I'd go that route because it is a very slick almost self-regulating way of moving the heat to be removed elsewhere. At current BTCvalue no way right now.