How is Epsilon 700W more powerful than Epsilon 600W at over 35 C???

Grace and all,

I keep reading information that implies that both the 600W and 700W power supplies perform the same at 35 C


I read in the X-bit labs review of the FSP Epsilon 700 W GLN that it had eight of Fairchild’s YM3045N (MBRP3045N) in small TO-220 packages which are very good.

Does the Epsilon 600W power supply have eight as well?

I could use the slightly longer cables of the 700W supply, but what is it in terms of the number and quality of the Fairchilds used that differentiates these to power supplies?

How and why would tempurature impact one more than the other?

It just seems that those that say the Epsilon 700W power supply drops 100W of its potential on its way from 25 C to 35 C are overstating the drop.

Please help me better understand this!

Mike

ITs because the efficiency fo the transistors drops off sharply as a function of temperature, so when they heat up more energy is lost and thus they dont perform as well.

Quote:

---

Originally Posted by herulach

ITs because the efficiency fo the transistors drops off sharply as a function of temperature, so when they heat up more energy is lost and thus they dont perform as well.

---

1) I kind of doubt the Epsilons efficiency drops off "sharply" at higher temps or higher loads!

2) This sort of reply is exactly why I am asking Grace for clarification

3) I have seen some surmise a 10W per 1C drop in potential power supplied, but Sparkle Power in its specs claims no more than a 5W drop.. and if I am not mistaken Sparkle uses the GLC model from FSP, which is certainly no better than the GLN.

4) I am fairly certain that the 700W model has a little more in it than the 600W.. I am merely inquiring as to what it is that makes the 700W unit better than the 600W. What are the differences between the units?

Quote:

---

Originally Posted by thegoldenstrand

1) I kind of doubt the Epsilons efficiency drops off "sharply" at higher temps or higher loads!

---

Well, then youd be wrong obviously, because thats what happens, hence the derating with temperature. More power generates substantialy more heat (IIRC its a square law)

Quote:

---

3) I have seen some surmise a 10W per 1C drop in potential power supplied, but Sparkle Power in its specs claims no more than a 5W drop.. and if I am not mistaken Sparkle uses the GLC model from FSP, which is certainly no better than the GLN.

---

Maybe, maybe not, but it could be better heatsinked, different fan design, or a plain mistake/blatantly false claim (it wouldnt be the first psu manufacturer to do so)

Quote:

---

4) I am fairly certain that the 700W model has a little more in it than the 600W.. I am merely inquiring as to what it is that makes the 700W unit better than the 600W. What are the differences between the units?

---

It does, i belive the transient currents are higher, as is the total sustainable load, providing its cooled properly.

Grace and all,

I am asking specifically about the two power supplies from FSP. My questions are not general, but specific in nature. Herulach has completely missed my point. I think we need to define the terms for Herulach as his use of the word "sharply" is not appropriate in this context unless he really believes that the 700 GLN's potential power drops more "sharply" than the 600 GLN model at higher tempuratures. Herulach is this what you are saying or were you not trying to really answer my question, but rather just trying to show off what you know?

Hence, I when I object to the use of the word "sharply" that Herulach used, it is in the context of comparing the two models the Epsilon 600GLN and 700 GLN.

Now.. compared to the 600 GLN does the "efficiency" or "potential power output" of the 700 GLN model drop sharply?

When i think of a "sharp" drop it is in the comparitive. For instance I read a study where the 680W Thermaltake Power Supply dropped more sharply than the PC Power & Cooling model. While we can split hairs and say both dropped "sharply", the fact remains that one dropped more "sharply" than the other. Of course this is way off topic, but comparitively the PC Power & Cooling unit dropped at a rate of 4 Watts per 1 C degree rise in Tempurature and the Thermaltake's potential power dropped about twice as fast.

Since my questions "obviously" related specifically to the two Epsilon Power Supplies, Is Herulach correct is stating that the 700 GLN model drops "sharply" compared to the 600 GLN model? and yes he did offend me.

He made a big point in correcting me and telling me I was obviously wrong when I objected to his statements. But my question was specific and I am not here to teach Herulach or anyone else common courtesy or how to read and write.

Mike

I think someone needs to chill out. I never meant to cause any offence, in the way i used i meant that if you were to look at a graph of power output vs temperature you would see a sharp drop in output as you increased the temperature, as you stated, nearly an amp per degree is quite a drop off, this is in comparison to what i would generally call a smooth derating from most supplies (and probably the rest of the curve for 700GLN, until you reach thermal breakdown obviously).

Now ive tried to answer your question, you dont like my answer fine, but dont get all narky about it, the best bit of advice would be if you want to know why (what i assume to be) MOSFETs behave like they do, then go read an EE book, or even use google, dontask on a forum.

Quote:

---

Originally Posted by thegoldenstrand

Grace and all,

I am asking specifically about the two power supplies from FSP. My questions are not general, but specific in nature. Herulach has completely missed my point. I think we need to define the terms for Herulach as his use of the word "sharply" is not appropriate in this context unless he really believes that the 700 GLN's potential power drops more "sharply" than the 600 GLN model at higher tempuratures. Herulach is this what you are saying or were you not trying to really answer my question, but rather just trying to show off what you know?

Hence, I when I object to the use of the word "sharply" that Herulach used, it is in the context of comparing the two models the Epsilon 600GLN and 700 GLN.

Now.. compared to the 600 GLN does the "efficiency" or "potential power output" of the 700 GLN model drop sharply?

When i think of a "sharp" drop it is in the comparitive. For instance I read a study where the 680W Thermaltake Power Supply dropped more sharply than the PC Power & Cooling model. While we can split hairs and say both dropped "sharply", the fact remains that one dropped more "sharply" than the other. Of course this is way off topic, but comparitively the PC Power & Cooling unit dropped at a rate of 4 Watts per 1 C degree rise in Tempurature and the Thermaltake's potential power dropped about twice as fast.

Since my questions "obviously" related specifically to the two Epsilon Power Supplies, Is Herulach correct is stating that the 700 GLN model drops "sharply" compared to the 600 GLN model? and yes he did offend me.

He made a big point in correcting me and telling me I was obviously wrong when I objected to his statements. But my question was specific and I am not here to teach Herulach or anyone else common courtesy or how to read and write.

Mike

---

Herulach,

If you understood what I was asking originally, it was never about power supplies losing power with tempurature or efficiency with tempurature. It was clearly how or why those like yourself could ever assert that the Epsilon 700 GLN did so in the same operating conditions at a faster rate than the Epsilon 600 GLN.

How and why would the operating tempurature impact the 700 GLN more than the 600 GLN is and was the call of the question. It called for a specific comparitive answer and you have so far failed miserably at even approaching the answer that was appropriate for the question. Do you want to try again?

I will not argue with you on the issue of whether or not you were or are trying. No doubt you are a trying individual.

Mike

Quote:

---

Originally Posted by thegoldenstrand

How and why would the operating tempurature impact the 700 GLN more than the 600 GLN is and was the call of the question. It called for a specific comparitive answer and you have so far failed miserably at even approaching the answer that was appropriate for the question. Do you want to try again?

---

How witty of you, the 700 is rated at 25 degrees, the 600 at 35, as a cursory glance at the spec sheet would reveal. So assuming a room temp of 25 degrees, and a generous 10 over ambient for the internals (it will almost certainly be more). The 700 is running ten degrees hotter than when rated, where as the 600 is running at the 'correct' temperature.

Herulach,

I have read this before. Just curious, but how do you believe this sort of answer remotely answers the question? Ratings at different tempuratures don't really tell you much. I have asked Grace for ratings at the same tempuratures. I personally abuse my computers temps and a more appropriate tempurature for me to compare these power supplies would be at least 40 C and in some instances, worst case, i would like to know what they are rated at when the tempurature rises to 50 C

Starting at the same tempurature, where both are rated, which power supply will see the "sharper" decline??

Mike

Good evening to both of you :-)
Erm, it does seem to have become a heated debate here... it reminds me more of a political discussion than one on PSU's and so i feel i should step in as i also know Grace is alittle snowed under with work right now :-)))
First of all i have to say it seems alittle extreme to say our Epsilon 700 would drop 100w from 25°c to 35°c (first time i heard that one), as i also know we are testing all (and i really mean all) of our PSU's to 50°c.
For the Epsilon 700, the continuous power rating at 40°c is 700w and at 50°c 650w (so from 40°c to 50°c it looses 5w per °c). So the idea of loosing already 100w between 25°c and 35°c is way off the mark... likewise the Epsilon 700 isn't going to be "that" more affected than the 600... they're both sturdy performers.
Hope the info helps ~ and i hope you both continue contributing :-)))))

Heidi,

I do thank you for stepping in. However, it was that spec sheet for the Epsilon FX-700 GLN that on page nine stated it was rated at 700W at 25 C and then provided information that assuming a 10W per C drop would give the Epsilon FX-700 GLN power supply a 600 W rating at 35 C and has led to more than a couple articles stating that the power of this supply was comparable to the FX-600 GLN at 35 C. I have looked at the spec sheets again and believe FSP may need to update them based upon what you have just posted. Please note I did not find any tempurature rating for the FX-600 GLN on its spec sheet.

Mike

http://www.home2000.net/client/fspgr...0FX700-GLN.pdf

see the top of page nine

Heidi or Grace,

What is the FX-600 GLN rated at for 40 C and 50 C?

thanks!

Mike

Hi Mike,

I was on the road a whole day yesterday so just read the messages above now.

I am glad that Heidi was helping to provide more info yesterday. I am afraid that I need to get this specific rating for you next week since it's now too late to talk to any engineer or PM in Taiwan due to the time difference. All the doc I have in hand can't provide more details on this.

I will send a mail to them now and see whether they can answer the question to you via email directly or I have to speak to them on Tue after Easter.

Grace

Hi Mike,

As my mail to you, I really appreciate your feedback on the incorrect spec you found from US.

I have just talked to our Product Manager in Taiwan about your question. The correct data is as below.

FX - 700GLN
40C - 700W
50C - 650W

FX - 600GLN
40C - 600W
50C - 600W

Hope this is helpful.
Grace

Grace,

Thank you!!