New Official AMT Thread

[IMET]

Just tossing this out there for general information:
http://www.edn.com/electronics-blogs/the-workbench/4421218/PCB-layout-tips-for-thermal-vias

http://circuitcalculator.com/wordpress/2006/03/12/pcb-via-calculator/

http://m.eet.com/media/1200036/XLamp_PCB_Thermal.pdf

Those are examples of standard industry practice reference items for thermal via design. Even if not explicitly specified in 'build-to-print' any CM should:
a. Knowing the thermal needs for the A1 assumed maximum thermal transfer is needed meaning; heavy-plated thermal vias at a minimum preferably filled as well, and a full complement of thermal contact bumps on the bottom of the board that are high enough to go past the gap created by green screen layer to make firm mechanical contact with the main heat sink. btw, that layer creates a 3mil air gap. If there are a few bumps but not a full complement - that gap just got increased to 8-10mil. Thermal compound is not going to make up for that...

b. Flagged the above concerns and asked their customer to verify what is needed to meet industry-standard practice for working with high-power packages. If the customer is unsure then follow Best Practice and overkill the issue.

Then there is the matter of basic Ohms law... If the value of 1 component in a divider network is changed while leaving all others the same - the result (output) from that network is vastly different. To keep the network balanced all components need to be scaled by the same percentage..

I will leave it up to the Peanut Gallery to figure how this applies to us here I will say that a power chip missing over 50% of its expected cooling path combined with getting hit (very briefly) with excessive core voltage is not long for this world... These are all items that any CM with a design team should have found in initial design review and later design change reviews and known how to address properly.

There is always room to optimize and info was presented to AMT (although the task was to build Bitmine designs "as is" because they had a proven working design); however, the boards can and do work fine "as is" when assembled properly from a mechanical/thermal standpoint (not PCB Assembly/soldering standpoint).  This is something that this CM was not responsible for as we did not have the capacity.

[1714834431]

1714834431

[1714834431]

1714834431

[NotFuzzyWarm]

aside from many/most of the backside thermal contact pads having anywhere from 25 to 50% of the thermal contact bumps not being there... Make nice jacks for pushing the board away from the sink.

The big silver dots are smushed (as they should be) thermal bumps. Most are not present at all. Folks who still have miners unreturned can easily unbolt the heatsinks and verify this. If the bumps were such a problem to make properly then even a thin copper foil shim would have done the trick.

[IMET]

aside from many/most of the backside thermal contact pads having anywhere from 25 to 50% of the thermal contact bumps not being there... Make nice jacks for pushing the board away from the sink.

Folks who still have miners unreturned can easily unbolt the heatsinks and verify this. If the bumps were such a problem to make properly then even a thin copper foil shim would have done the trick.

But when mechanicals/thermals are not assembled properly and you just keep putting power on unit after unit that was delivered working, you end up with a huge scrap pile. 

[NotFuzzyWarm]

aside from many/most of the backside thermal contact pads having anywhere from 25 to 50% of the thermal contact bumps not being there... Make nice jacks for pushing the board away from the sink.

Folks who still have miners unreturned can easily unbolt the heatsinks and verify this. If the bumps were such a problem to make properly then even a thin copper foil shim would have done the trick.

But when mechanicals/thermals are not assembled properly and you just keep putting power on unit after unit that was delivered working, you end up with a huge scrap pile.  

No argument there on the scrap factor. Thermal bumps are not final assembly. They are either on the board at time of production or not. And yes  continuing to bolt on sinks and burn, er, run the boards when they have such an easily seen by eyeball & felt by finger defect certainly not very smart.

[FrictionlessCoin]

Just tossing this out there for general information:
http://www.edn.com/electronics-blogs/the-workbench/4421218/PCB-layout-tips-for-thermal-vias

http://circuitcalculator.com/wordpress/2006/03/12/pcb-via-calculator/

http://m.eet.com/media/1200036/XLamp_PCB_Thermal.pdf

Those are examples of standard industry practice reference items for thermal via design. Even if not explicitly specified in 'build-to-print' any CM should:
a. Knowing the thermal needs for the A1 assumed maximum thermal transfer is needed meaning; heavy-plated thermal vias at a minimum preferably filled as well, and a full complement of thermal contact bumps on the bottom of the board that are high enough to go past the gap created by green screen layer to make firm mechanical contact with the main heat sink. btw, that layer creates a 3mil air gap. If there are a few bumps but not a full complement - that gap just got increased to 8-10mil. Thermal compound is not going to make up for that...

b. Flagged the above concerns and asked their customer to verify what is needed to meet industry-standard practice for working with high-power packages. If the customer is unsure then follow Best Practice and overkill the issue.

Then there is the matter of basic Ohms law... If the value of 1 component in a divider network is changed while leaving all others the same - the result (output) from that network is vastly different. To keep the network balanced all components need to be scaled by the same percentage..

I will leave it up to the Peanut Gallery to figure how this applies to us here I will say that a power chip missing over 50% of its expected cooling path combined with getting hit (very briefly) with excessive core voltage is not long for this world... These are all items that any CM with a design team should have found in initial design review and later design change reviews and known how to address properly.

There is always room to optimize and info was presented to AMT (although the task was to build Bitmine designs "as is" because they had a proven working design); however, the boards can and do work fine "as is" when assembled properly from a mechanical/thermal standpoint (not PCB Assembly/soldering standpoint).  This is something that this CM was not responsible for as we did not have the capacity.

Indeed interesting that these were the exact Bitmine designs.  The Bitmine designs did have flaws as revealed by their late deliveries.  However,  Bitmine actually did deliver working units.  Not the same case as AMT.

Still let's get back to reality,  AMT has never made an effort to fix customer issues.  I got a system that did not include a backplane.  AMT never made the effort to ship a replacement. 

[NotFuzzyWarm]

No argument there. BM.ch seems to have recovered from those mistakes much better.

[FrictionlessCoin]

No argument there. BM.ch seems to have recovered from those mistakes much better.

Recovered because they make a serious effort.   AMT has not made a smidgen of an effort.

Just look at their RMA process that's going to some parking lot behind a Mason temple.  It's ridiculous.

[ladi4ever]

Post deleted, because we are hurting this company so much, this company that was created from a scam er mastermind ..

[raveneffect2333]

Posting here again on an update of my situation. Received an email from AMT stating my delivery would be "Monday" this was on the 15th of June. Also offered me a hosted solution at this time.

Its two weeks later and 4 emails sent by myself. It appears they have stopped talking to me again. No Miners, No updates, No tracking numbers, No replies and No hosted solution. Definitely no refunds.

We are now heading towards 8 months since payment was sent to you guys. There is no answer for this besides "sorry we stuffed up, here is your money back"

[rik_khaos]

So if you got a petahash of hosting, and you are selling it at $1 per Ghash you are looking at 1mil.

It's rough to trust you while being burned so far, but there is no reason that your previous customers should not be able to trade their current order for hosted mining by turning every dollar spent into a GH of hashing power.

I spent $5,635 ( I took of my express shipping there) so I would rather the money I spent with you be transferred to hosting option and receive 5.6TH/s  I would consider that reasonable  and fair  to my self as a consumer and to you as a company and would be in line with what would be needed for mpp.

[opieum2]

So if you got a petahash of hosting, and you are selling it at $1 per Ghash you are looking at 1mil.

It's rough to trust you while being burned so far, but there is no reason that your previous customers should not be able to trade their current order for hosted mining by turning every dollar spent into a GH of hashing power.

I spent $5,635 ( I took of my express shipping there) so I would rather the money I spent with you be transferred to hosting option and receive 5.6TH/s  I would consider that reasonable  and fair  to my self as a consumer and to you as a company and would be in line with what would be needed for mpp.

And considering if it was at that rate, it would be good. I personally am more interested in getting the hardware miners, with a similar hashrate. But either way if they do what they need to to honor it then it will work out.

[raydeeyo]

Haven't read AMT thread in a long time since the previous one was closed, thought not to ask in here but after 2 month of mailing & calling and yet to get any replies... here is the last place i can go... So did anyone receive their miners yet? When was it ya? Can i know your order number so i can know that it's still operating... mine was #1433

AMT, y u ignore my question...

We were working on the last post, sry.  Yes we cannot discuss clients individual orders on this public forum. someone will get back to you via email shortly, thanks.

Hi AMT, no one get back to me yet, can you help to check?

Day 3, i really need to talk to you, would you help to reply my mail? Order #1433

AMT, please reply my mail. 'TQ

AMT, if you were here, please reply my mail. Order #1433

[FrictionlessCoin]

So if you got a petahash of hosting, and you are selling it at $1 per Ghash you are looking at 1mil.

It's rough to trust you while being burned so far, but there is no reason that your previous customers should not be able to trade their current order for hosted mining by turning every dollar spent into a GH of hashing power.

I spent $5,635 ( I took of my express shipping there) so I would rather the money I spent with you be transferred to hosting option and receive 5.6TH/s  I would consider that reasonable  and fair  to my self as a consumer and to you as a company and would be in line with what would be needed for mpp.

And considering if it was at that rate, it would be good. I personally am more interested in getting the hardware miners, with a similar hashrate. But either way if they do what they need to to honor it then it will work out.

The reason I mistakenly selected AMT was because all I thought they were doing was assembling miners.  All the technology was sourced elsewhere.  How difficult can this be?

Now they are claiming that they want to do something much more difficult.  Hosted mining, when they can't even do simple assembly?

AMT as a company is pretty much done.  There's really no reputation to salvage.  

[AMT_miners]

As we previously stated the hosted farm is Bitmine's and we're collaborating as usual. If you're interested in the farm you can purchase through us or directly through bitmine at the same price. Down the road we will contribute to the farms second hashing power addition which will be in August - September with bitmine together, and we'll  be helping them sell as well.

[C0Kez3R0]

I haven't checked in here for a while. Just wondering when I'm going to get my $6000+ back. To any who have been keeping up, are we any closer to a final resolution to this situation?

[mrpark]

I click on bitmine.ch about cloud hashing. It says $2,026 a year for a 1T contract. Can you explain how is this works out to $1/gig?

[YourPalToots]

That's great that you now have cloud hashing.

Instead of shipping this rock back and forth you can give me 6.159T of hashing.

[FrictionlessCoin]

That's great that you now have cloud hashing.

Instead of shipping this rock back and forth you can give me 6.159T of hashing.

So has the RMA process ever worked?  Did you get a new machine for the broken machine you sent to them?

[YourPalToots]

That's great that you now have cloud hashing.

Instead of shipping this rock back and forth you can give me 6.159T of hashing.

So has the RMA process ever worked?  Did you get a new machine for the broken machine you sent to them?

My RMA delivery was refused by AMT and sent back.

[dreward]

But you said you had filed in the court of common pleas. That is why I am confused