If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Rate Thread | Display Modes |
#1
|
|||
|
|||
corsair RM750i question
This PSU has a feature to enable single 12 volt rail at 62.5 amps or
multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? Rene |
Ads |
#2
|
|||
|
|||
corsair RM750i question
Rene Lamontagne wrote:
This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. 5 rails at 40A Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? |
#3
|
|||
|
|||
corsair RM750i question
On 01/11/2018 11:36 AM, Andy Burns wrote:
Rene Lamontagne wrote: Â*Â* This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. 5 rails at 40A Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? Yeah, your right. looked at Corsairs site now instead of reviewrs. Boy that's even worse math. 5 x 40 = 200 amps !!! Rene |
#4
|
|||
|
|||
corsair RM750i question
Rene Lamontagne wrote:
This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab. There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU. The 40A is peak load, not sustained load nor independent load. The /combined/ load across all those 12V rails will still be a max of 750W. So, 750W / 12V = 62.5A. You're still going to hit the COMBINED LIMIT of amperage across all the rails. You are NOT going to get 2400W @ 12V across all those 4 rails. Doesn't matter if you use a single rail or multiple rails. The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use. This is not new regarding PSU specifications. You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A |
#5
|
|||
|
|||
corsair RM750i question
Rene Lamontagne wrote:
looked at Corsairs site now instead of reviewrs. Boy that's even worse math. 5 x 40 = 200 amps !!! Yes,but as VanguardLH points out, it does show the maximum combined wattage for the various rails. |
#6
|
|||
|
|||
corsair RM750i question
On 01/11/2018 12:45 PM, VanguardLH wrote:
Rene Lamontagne wrote: This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab. There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU. The 40A is peak load, not sustained load nor independent load. The /combined/ load across all those 12V rails will still be a max of 750W. So, 750W / 12V = 62.5A. You're still going to hit the COMBINED LIMIT of amperage across all the rails. You are NOT going to get 2400W @ 12V across all those 4 rails. Doesn't matter if you use a single rail or multiple rails. The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use. This is not new regarding PSU specifications. You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A Yes VangardLH, I know all that, And you know all that I've been in the Electrical/electronics field for the last 60 years, It just strikes me funny the way its set out, They sorta lead the uninitiated people to believe there's more there than there actually is, They could do a footnote "to a maximum of 62.5 amps from all 5 rails" Rene |
#7
|
|||
|
|||
corsair RM750i question
On 01/11/2018 1:04 PM, Rene Lamontagne wrote:
On 01/11/2018 12:45 PM, VanguardLH wrote: Rene Lamontagne wrote: This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab.Â* There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU.Â* The 40A is peak load, not sustained load nor independent load.Â* The /combined/ load across all those 12V rails will still be a max of 750W.Â* So, 750W / 12V = 62.5A.Â* You're still going to hit the COMBINED LIMIT of amperage across all the rails.Â* You are NOT going to get 2400W @ 12V across all those 4 rails.Â* Doesn't matter if you use a single rail or multiple rails.Â* The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use.Â* This is not new regarding PSU specifications.Â* You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A Yes VangardLH, I know all that, And you know all that I've been in the Electrical/electronics field for the last 60 years, It just strikes me funny the way its set out, They sorta lead the uninitiated people to believe there's more there than there actually is, They could do a footnote "to a maximum of 62.5 amps from all 5 rails" Rene On reading it closer I guess they do actually say that. Rene |
#8
|
|||
|
|||
corsair RM750i question
Rene Lamontagne wrote:
On 01/11/2018 12:45 PM, VanguardLH wrote: Rene Lamontagne wrote: This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab. There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU. The 40A is peak load, not sustained load nor independent load. The /combined/ load across all those 12V rails will still be a max of 750W. So, 750W / 12V = 62.5A. You're still going to hit the COMBINED LIMIT of amperage across all the rails. You are NOT going to get 2400W @ 12V across all those 4 rails. Doesn't matter if you use a single rail or multiple rails. The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use. This is not new regarding PSU specifications. You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A Yes VangardLH, I know all that, And you know all that I've been in the Electrical/electronics field for the last 60 years, It just strikes me funny the way its set out, They sorta lead the uninitiated people to believe there's more there than there actually is, They could do a footnote "to a maximum of 62.5 amps from all 5 rails" They probably figure consumers that are buying PSUs are educated. Those that aren't should be buying pre-built computers and having shops do the repairs. Would you dismantle your car's engine to replace its intake manifold filter or replace the hydraulic lifters for the valves without being educated how to do so? Nothing to stop you from putzing around on your car's engine but being uneducated means not doing the job right. Would you sneak into an operating room pretending to be a qualified medical technician in charge of a blood transfusion? The PSU is the life's blood of a computer. They expect to sell PSUs to the educated - technical folks that job their own computers by building from parts. If you're parting out your own computer job and doing the build yourself, you should already know about the aggregate loading across multiple rails. And, as you noted, the math is simple (for DC): volts times amperes equals wattage. Even the PSU-uneducated can do the simple math to see the total of five rails at 484 watts would total 2420 watts - and obviously even those uneducated users know they are not paying for a 2400+ watt PSU (assuming they could even find one). It takes but once seeing the simple math doesn't add up to initiate a search to educate. When has Marketing ever been accountable to the absolute truth? You and I count multiples of bytes in powers of 2. You and I know that drives are sold based in powers of 10 regarding capacity so the drives look larger to consumers. Your original post did not address the issue to the mass of uberboobs using computers. You only addressed it for yourself. I wasn't sure why this was magic to you. Mostly I figured you miscounted the 12V rail count and was opulent in my verbose reply by adding more than just a rail miscount (I don't mentally track the personal life experiences of those to whom I reply so the added information was "just in case you didn't know"). |
#9
|
|||
|
|||
corsair RM750i question
Rene Lamontagne wrote:
On 01/11/2018 1:04 PM, Rene Lamontagne wrote: On 01/11/2018 12:45 PM, VanguardLH wrote: Rene Lamontagne wrote: This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab. There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU. The 40A is peak load, not sustained load nor independent load. The /combined/ load across all those 12V rails will still be a max of 750W. So, 750W / 12V = 62.5A. You're still going to hit the COMBINED LIMIT of amperage across all the rails. You are NOT going to get 2400W @ 12V across all those 4 rails. Doesn't matter if you use a single rail or multiple rails. The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use. This is not new regarding PSU specifications. You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A Yes VangardLH, I know all that, And you know all that I've been in the Electrical/electronics field for the last 60 years, It just strikes me funny the way its set out, They sorta lead the uninitiated people to believe there's more there than there actually is, They could do a footnote "to a maximum of 62.5 amps from all 5 rails" Rene On reading it closer I guess they do actually say that. Rene There are two aspects to the design: 1) Actual components that output the power (12V @ 62A say). 2) Various overcurrent features exist, to keep the PSU design safe for end-user usage. The overcurrent detector turns off the supply, for whatever combo of detectors you put in place. Some spec (IEC 950? or UL 60950?) defines SELV. That has two aspects. What DC voltage is too high for safety (-48V?). What total power on a loom set is too much (12V @ 20A? limit). In the case of your stated supply, what they're doing is switching off the 40A per-loom detectors, and leaving the 62A overall (probably thermally limited) output capability. When the 40A detectors are enabled, it still only makes 62A max total, but it will switch off the supply if a subtending set of wires hits 40A. It's probably still not consistent with SELV, in either mode. SELV predicts around 20A on a loom set, as safe. ******* A quick SELV check... https://www.edn.com/electronics-blog...power-supplies "Most switchmode low voltage AC-DC power supplies with outputs up to 48VDC meet the SELV requirements." [on hazardous voltage] All this 48V baloney comes from telecom and telco boat-anchor power supplies (or battery rooms). And the power limit, that defines the amps, consists of some curves. I can see where the old 240W idea came from. 30V at 8 amps. And a second diagram predicts other sets of conditions. http://acstestlab.com/understanding-...rements-part-3 "Table 2C outlines the limits for overcurrent device limited sources." At 12V, the graph says (roughly): 12V @ ~20A max under any load except short circuit 12V @ ~90A max under short circuit 12V @ 130A max for "circuit breaker rating" The PSU doesn't have a circuit breaker. It simply stops the switching of the primary side switching transistors, to cause power to drop to zero. So the third item doesn't apply. And the second item is met by the current measurement schemes (40A or 62A or whatever). However, the supply is not correctly designed for the (kinda) wire-induced or transformer-induced limitations. You should break it up into looms of no more than 20A each. Which means doubling the number of current monitor circuits ($$$). So the supply may still not be totally consistent with 60950. It's in a metal box, which I guess counts as a fire enclosure, except when a flame shoots out where the exhaust fan is located (this has happened... but not in modern times). The design of power supplies still bothers me a bit, and I've yet to see a "clever justification" for the feature set. People seem to like to cut corners on SELV, so they must have some "not guilty, Your Honor" idea in mind when a product liability suit comes up in court. Where I worked, we did a full UL workup. It costs a lot of money to do that (like burning an entire system, in a UL fire facility). They burn your goods, and video it, so your mechanical designer can redesign a few pieceparts to "fix" it. No matter what happens, when you stuff burns, it's not supposed to "encourage" an inferno. It's supposed to snuff out or smoke, not open flame. There were only a few power supplies, ones with a really long chassis (hits your CD drive), where the four looms came from four separate transformers. And that doesn't mean a thing to SELV. Doesn't help. The thing still has to current limit, on a per-loom basis, in such a way as to be compliant with some spec. Paul |
#10
|
|||
|
|||
corsair RM750i question
On 01/11/2018 3:55 PM, Paul wrote:
Rene Lamontagne wrote: On 01/11/2018 1:04 PM, Rene Lamontagne wrote: On 01/11/2018 12:45 PM, VanguardLH wrote: Rene Lamontagne wrote: This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab.Â* There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU.Â* The 40A is peak load, not sustained load nor independent load.Â* The /combined/ load across all those 12V rails will still be a max of 750W.Â* So, 750W / 12V = 62.5A.Â* You're still going to hit the COMBINED LIMIT of amperage across all the rails.Â* You are NOT going to get 2400W @ 12V across all those 4 rails.Â* Doesn't matter if you use a single rail or multiple rails.Â* The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use.Â* This is not new regarding PSU specifications.Â* You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A Yes VangardLH, I know all that, And you know all that I've been in the Electrical/electronics field for the last 60 years, It just strikes me funny the way its set out, They sorta lead the uninitiated people to believe there's more there than there actually is, They could do a footnote "to a maximum of 62.5 amps from all 5 rails" Rene On reading it closer I guess they do actually say that. Rene There are two aspects to the design: 1) Actual components that output the power (12V @ 62A say). 2) Various overcurrent features exist, to keep the PSU Â*Â* design safe for end-user usage. The overcurrent detector Â*Â* turns off the supply, for whatever combo of detectors Â*Â* you put in place. Some spec (IEC 950? or UL 60950?) defines SELV. That has two aspects. What DC voltage is too high for safety (-48V?). What total power on a loom set is too much (12V @ 20A? limit). In the case of your stated supply, what they're doing is switching off the 40A per-loom detectors, and leaving the 62A overall (probably thermally limited) output capability. When the 40A detectors are enabled, it still only makes 62A max total, but it will switch off the supply if a subtending set of wires hits 40A. It's probably still not consistent with SELV, in either mode. SELV predicts around 20A on a loom set, as safe. ******* A quick SELV check... https://www.edn.com/electronics-blog...power-supplies Â*Â* "Most switchmode low voltage AC-DC power supplies with Â*Â*Â* outputs up to 48VDC meet the SELV requirements."Â*Â*Â*Â*Â*Â* [on hazardous voltage] All this 48V baloney comes from telecom and telco boat-anchor power supplies (or battery rooms). And the power limit, that defines the amps, consists of some curves. I can see where the old 240W idea came from. 30V at 8 amps. And a second diagram predicts other sets of conditions. http://acstestlab.com/understanding-...rements-part-3 Â*Â* "Table 2C outlines the limits for overcurrent device limited sources." Â*Â* At 12V, the graph says (roughly): Â*Â* 12V @ ~20AÂ* max under any load except short circuit Â*Â* 12V @ ~90AÂ* max under short circuit Â*Â* 12V @ 130AÂ* max for "circuit breaker rating" The PSU doesn't have a circuit breaker. It simply stops the switching of the primary side switching transistors, to cause power to drop to zero. So the third item doesn't apply. And the second item is met by the current measurement schemes (40A or 62A or whatever). However, the supply is not correctly designed for the (kinda) wire-induced or transformer-induced limitations. You should break it up into looms of no more than 20A each. Which means doubling the number of current monitor circuits ($$$). So the supply may still not be totally consistent with 60950. It's in a metal box, which I guess counts as a fire enclosure, except when a flame shoots out where the exhaust fan is located (this has happened... but not in modern times). The design of power supplies still bothers me a bit, and I've yet to see a "clever justification" for the feature set. People seem to like to cut corners on SELV, so they must have some "not guilty, Your Honor" idea in mind when a product liability suit comes up in court. Where I worked, we did a full UL workup. It costs a lot of money to do that (like burning an entire system, in a UL fire facility). They burn your goods, and video it, so your mechanical designer can redesign a few pieceparts to "fix" it. No matter what happens, when you stuff burns, it's not supposed to "encourage" an inferno. It's supposed to snuff out or smoke, not open flame. There were only a few power supplies, ones with a really long chassis (hits your CD drive), where the four looms came from four separate transformers. And that doesn't mean a thing to SELV. Doesn't help. The thing still has to current limit, on a per-loom basis, in such a way as to be compliant with some spec. Â*Â* Paul Thanks Paul, I always appreciate your concise explanations. The specs do lead me to believe that it is well protected against nearly all voltage and current problems, so I am OK on that. Would be nice to see some schematics of some of these modern supplies, All I find are on the older models. Rene |
#11
|
|||
|
|||
corsair RM750i question
On Thu, 11 Jan 2018 16:21:42 -0600, Rene Lamontagne
wrote: Thanks Paul, I always appreciate your concise explanations. The specs do lead me to believe that it is well protected against nearly all voltage and current problems, so I am OK on that. Would be nice to see some schematics of some of these modern supplies, All I find are on the older models. http://www.overclock.net/t/761202/si...rail-explained |
#12
|
|||
|
|||
corsair RM750i question
wrote:
On Thu, 11 Jan 2018 16:21:42 -0600, Rene Lamontagne wrote: Thanks Paul, I always appreciate your concise explanations. The specs do lead me to believe that it is well protected against nearly all voltage and current problems, so I am OK on that. Would be nice to see some schematics of some of these modern supplies, All I find are on the older models. http://www.overclock.net/t/761202/si...rail-explained When you look at the reviews here, they have pictures, and you can count transformers on the "true multirail" ones. http://www.jonnyguru.com/modules.php...=Story&reid=15 "Below we see the two 12V transformers that are eventually split up into five 12V rails." And the label on that supply, actually reflects the internal construction :-) It's a miracle. http://www.jonnyguru.com/modules/NDR...K/IMG_0802.jpg HTH, Paul |
#14
|
|||
|
|||
corsair RM750i question,
On 01/11/2018 1:44 PM, VanguardLH wrote:
Rene Lamontagne wrote: On 01/11/2018 12:45 PM, VanguardLH wrote: Rene Lamontagne wrote: This PSU has a feature to enable single 12 volt rail at 62.5 amps or multiple, 4 rails at 40 amps. Now when I went to school this added up to 160 amps! Is this our new math or are Corsair fudging things? http://www.corsair.com/en-us/rmi-ser...ly-modular-psu Click on the "TECH SPECS" tab. There are *5* 12-volt rails in multiple rail mode, not 4. 5 * 40A * 12V = 2400W Obviously you aren't paying or getting a 2400 watt PSU. The 40A is peak load, not sustained load nor independent load. The /combined/ load across all those 12V rails will still be a max of 750W. So, 750W / 12V = 62.5A. You're still going to hit the COMBINED LIMIT of amperage across all the rails. You are NOT going to get 2400W @ 12V across all those 4 rails. Doesn't matter if you use a single rail or multiple rails. The combined power output of one or 5 of them is still the same. If one of those multi-rail outputs sucked down 40A (god forbid), that leaves only 22.5A total remaining for the other rails to use. This is not new regarding PSU specifications. You don't just add up the total wattage of every output to figure out what each voltage can supply altogether. https://www.youtube.com/watch?v=rC-ImSeYn2A Yes VangardLH, I know all that, And you know all that I've been in the Electrical/electronics field for the last 60 years, It just strikes me funny the way its set out, They sorta lead the uninitiated people to believe there's more there than there actually is, They could do a footnote "to a maximum of 62.5 amps from all 5 rails" They probably figure consumers that are buying PSUs are educated. Those that aren't should be buying pre-built computers and having shops do the repairs. Would you dismantle your car's engine to replace its intake manifold filter or replace the hydraulic lifters for the valves without being educated how to do so? Nothing to stop you from putzing around on your car's engine but being uneducated means not doing the job right. Would you sneak into an operating room pretending to be a qualified medical technician in charge of a blood transfusion? The PSU is the life's blood of a computer. They expect to sell PSUs to the educated - technical folks that job their own computers by building from parts. If you're parting out your own computer job and doing the build yourself, you should already know about the aggregate loading across multiple rails. And, as you noted, the math is simple (for DC): volts times amperes equals wattage. Even the PSU-uneducated can do the simple math to see the total of five rails at 484 watts would total 2420 watts - and obviously even those uneducated users know they are not paying for a 2400+ watt PSU (assuming they could even find one). It takes but once seeing the simple math doesn't add up to initiate a search to educate. When has Marketing ever been accountable to the absolute truth? You and I count multiples of bytes in powers of 2. You and I know that drives are sold based in powers of 10 regarding capacity so the drives look larger to consumers. Your original post did not address the issue to the mass of uberboobs using computers. You only addressed it for yourself. I wasn't sure why this was magic to you. Mostly I figured you miscounted the 12V rail count and was opulent in my verbose reply by adding more than just a rail miscount (I don't mentally track the personal life experiences of those to whom I reply so the added information was "just in case you didn't know"). Just a followup and Info. After 5 days and being on prime 2 day delivery I still had no word from Amazon, So checked my account and it said Not shipped yet` Did a check on the unit and their is no out of stock on it BUT the price has increased by $7.00 since I ordered it!!! So I cancelled the order and ordered one from Newegg.cafor $32.00 less! Should have looked there first. Rene |
#15
|
|||
|
|||
corsair RM750i question,
Rene Lamontagne wrote:
Just a followup and Info. After 5 days and being on prime 2 day delivery I still had no word from Amazon, So checked my account and it said Not shipped yet` Did a check on the unit and their is no out of stock on it BUT the price has increased by $7.00 since I ordered it!!! So I cancelled the order and ordered one from Newegg.cafor $32.00 less! Should have looked there first. Rene I had an experience with Amazon like that too, which is why Amazon is on my "never again" list. They put the charge on the CC and everything, then decided they had no stock. Took a month to get the refund on the CC. The "replacement item", same make and model, they were willing to offer me for "full retail". What a bargain. Bait and switch. Yarrh. And on my first attempted transaction. It's funny, my regular e-tailers don't do that to me. I guess that's why they're my regulars. Paul |
|
Thread Tools | |
Display Modes | Rate This Thread |
|
|