• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

FSP Readies a Redundant PSU for Standard ATX Desktops

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,681 (7.42/day)
Location
Dublin, Ireland
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard Gigabyte B550 AORUS Elite V2
Cooling DeepCool Gammax L240 V2
Memory 2x 16GB DDR4-3200
Video Card(s) Galax RTX 4070 Ti EX
Storage Samsung 990 1TB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
Fortron's channel brand FSP is ready with a unique new redundant power supply for standard ATX desktops. The new Twins Series PSU is an ATX-size box that encloses two hot-swappable 500W server-grade PSUs, with internal circuitry that maintains redundancy. The PSU further interfaces with your OS, giving you control over redundancy, monitoring, and programmable alerts. The PSU is ideal for home-servers and workstations with <500W power draw. Each of the two included subunits features 40 mm server-grade fans. FSP plans to reveal more information at its 2016 Computex exhibit.



View at TechPowerUp Main Site
 
Last edited by a moderator:
Nice idea, but I feel retaining one large fan in the housing is better than one small fan per unit.
 
Nice idea, but I feel retaining one large fan in the housing is better than one small fan per unit.

I think you missed the point of "redundancy." This is for mission critical servers. What happens if that one fan fails? What good do dual PSUs do you if they burn up and die?
 
Last edited:
No point missed, I've never had a 12cm or larger fan die in a PSU yet, but IMO a larger fan will keep it cooler and extend the life of both PSU's.
 
No point missed, I've never had a 12cm or larger fan die in a PSU yet, but IMO a larger fan will keep it cooler and extend the life of both PSU's.
Anecdotal evidence isn't enough to bet your servers or mission critical components on.
 
I think you missed the point of "redundancy." This is for mission critical servers. What happens if that one fan fails? What good to dual PSUs do you if they burn up and die?
Mission-critical servers does not use ATX-power supplies (really, really bad idea). But I get your point.
 
If it is reasonably priced, color me interested.

1U-4U server chassis are ridiculously expensive.
 
Mission-critical servers does not use ATX-power supplies (really, really bad idea). But I get your point.

The whole point of this product is to shatter that stereotype.
 
The whole point of this product is to shatter that conception.
This product is for single standalone servers, where there is not room for anything else. Just to add an extra bit of redundancy. It is absolutely not viable for mission critical use.
 
This product is for single standalone servers, where there is not room for anything else. Just to add an extra bit of redundancy. It is absolutely not viable for mission critical use.

I suppose it depends on what you define as "mission-critical"

My family pictures are critical to my "mission" as "keeper of the data." That said, I'd never buy a 1U chassis for them. I might consider this for a simple ATX backup server though.

I see where you are coming from though...
 
I suppose it depends on what you define as "mission-critical"

My family pictures are critical to my "mission" as "keeper of the data." That said, I'd never buy a 1U chassis for them. I might consider this for a simple ATX backup server though.

I see where you are coming from though...
Well, i suppose my definition of mission-critical is from my work. Machines that would cost tens of thousand of dollars per hour of downtime.
 
Well, i suppose my definition of mission-critical is from my work. Machines that would cost tens of thousand of dollars per hour of downtime.

This is really the only definition IMO, depending a bit.
 
This is really the only definition IMO, depending a bit.

Fair enough. I'm just being overly literal and taking words apart again, ignore it hehe.
 
I feel that we need an ATX PSU with integrated Li-Ion battery and UPS functionality... and with the ability to connect external li-ion pack for extra juice when needed...
 
Well, i suppose my definition of mission-critical is from my work. Machines that would cost tens of thousand of dollars per hour of downtime.

Yeah you seen how mission critical people get over gaming, and some of those CS ( and such ) contests can put you out lots of money too.
 
I feel that we need an ATX PSU with integrated Li-Ion battery and UPS functionality... and with the ability to connect external li-ion pack for extra juice when needed...
Or, you know, just buy a UPS. They dont cost that much.

Especially considering that ATX PSUS can remain in use for more than half a decade, you'd have to make a removable li-ion pack. PSUs are already highly packed with electrical components, a PSU with a UPS and sufficient battery size would be incredibly expensive, not to mention much bigger then the ATX standard.
 
Last edited:
I feel that we need an ATX PSU with integrated Li-Ion battery and UPS functionality... and with the ability to connect external li-ion pack for extra juice when needed...
li-ion? no thx, that would suck in various ways.
ni-mh? sure, trickle charging is great.
 
UPS use sealed lead acid batteries. They have a ~2 year life span and they're usually as big or bigger than the PSU.

I would hate replacing a battery inside of a UPS because to do it properly, you should disconnect the battery from the mains and the device it powers before extracting the battery itself. That would be a lot of work.

Not to mention, batteries get hot under load and adding that heat to the heat of the PSU itself does not add up to anything good.
 
Well, i suppose my definition of mission-critical is from my work. Machines that would cost tens of thousand of dollars per hour of downtime.
If you're earning that much, you're going to care a whole lot less about the reliability of a single component and would have architected it to be highly available. So if a server goes down, it's not the end of the world. So while I understand what you're saying, a single box should never be a single point of failure in such a system. At work we could lose possibly up to 10k an hour if our system goes down during peak hours, which is why we use cloud servers at Google so we can easily kill a server if it dies or is significantly acting up and spin up a new one. Our awesome sysadmin wrote Ansible scripts to automate all of that, so if the database server blows up, we can literally attach SAN storage and spin up a new box in 5 minutes with the database and all of its data ready to roll.

tl;dr: If your entire system earning you thousands an hour relies on a single box regardless of the redundancy inside the server itself, I would argue you've already made a terrible mistake.
 
If you're earning that much, you're going to care a whole lot less about the reliability of a single component and would have architected it to be highly available. So if a server goes down, it's not the end of the world. So while I understand what you're saying, a single box should never be a single point of failure in such a system. At work we could lose possibly up to 10k an hour if our system goes down during peak hours, which is why we use cloud servers at Google so we can easily kill a server if it dies or is significantly acting up and spin up a new one. Our awesome sysadmin wrote Ansible scripts to automate all of that, so if the database server blows up, we can literally attach SAN storage and spin up a new box in 5 minutes with the database and all of its data ready to roll.

tl;dr: If your entire system earning you thousands an hour relies on a single box regardless of the redundancy inside the server itself, I would argue you've already made a terrible mistake.
Yeah, but that was not the point. Your building it to prevent such an event (or any event for that matter), which is why it is called mission-critical.

You can't always "azure-up" the server if it borks out. We have certain machines that we absolutely cannot host in the cloud.
 
Nice! Someone was asking for it here on TPU just a few months ago, so if the price is right - there will definitely be a market of at least 2 people (myself included).

I feel that we need an ATX PSU with integrated Li-Ion battery and UPS functionality... and with the ability to connect external li-ion pack for extra juice when needed...
So far I've only seen PicoPSU variations with LiPo backup. A regular desktop PC needs something bigger, like a time proven dedicated UPS unit with lead-acid batteries o_O

This product is for single standalone servers, where there is not room for anything else. Just to add an extra bit of redundancy. It is absolutely not viable for mission critical use.
Why not? My neighbor's server room has one 2U rackmount and 2 ATX Full-towers. While it is easy to find a redundant PSU for that rack, we don't have any options for the other 2 servers.
 
Why not? My neighbor's server room has one 2U rackmount and 2 ATX Full-towers. While it is easy to find a redundant PSU for that rack, we don't have any options for the other 2 servers.
Well, thats the point. You don't use towers if your are a company that atleast have an IT Department.
At home it is great, in a work environment? I for sure can't see it.
 
If you're earning that much, you're going to care a whole lot less about the reliability of a single component and would have architected it to be highly available. So if a server goes down, it's not the end of the world. So while I understand what you're saying, a single box should never be a single point of failure in such a system. At work we could lose possibly up to 10k an hour if our system goes down during peak hours, which is why we use cloud servers at Google so we can easily kill a server if it dies or is significantly acting up and spin up a new one. Our awesome sysadmin wrote Ansible scripts to automate all of that, so if the database server blows up, we can literally attach SAN storage and spin up a new box in 5 minutes with the database and all of its data ready to roll.

tl;dr: If your entire system earning you thousands an hour relies on a single box regardless of the redundancy inside the server itself, I would argue you've already made a terrible mistake.


Lets not get too far off the reservation, this is intended as a "HOME"-server PSU, which usually are single box servers or even used in a small NAS build. So if some "IT guy" uses this for his companies server farm, I'd be worried.

As far as cooling goes (@Caring1 ) I only agree that a single 40mm fan might be a little small for each unit. However, one might mod it by add a large fan strategically placed, cant really tell from that one image alone, but it looks like the sides and top lack air holes.
 
Well, thats the point. You don't use towers if your are a company that atleast have an IT Department.
At home it is great, in a work environment? I for sure can't see it.
Depends on the demands of the company. I could see a tower chassis easily being a backup gateway server that's on a rack. The backup server is almost never running so, you don't want to take up rack space but, on the same token, you want to have a backup in case something happens to the real thing.
 
Lets not get too far off the reservation, this is intended as a "HOME"-server PSU
Say what? It's a 500W redundant unit. It can easily run a dual-socket server, or a decent workstation. It's definitely not aimed at consumer market. The only alternative on the market today is this $450 PoS which has been on the market for quite a few years:
http://www.newegg.com/Product/Produ...redundant_power_supply-_-17-338-047-_-Product

If someone uses Athenapower in their system - that's when you should be worried.
 
Back
Top