The power supply is heating up, is the video card guilty?

nicebloodx
And blow off the dust - a compressor or a vacuum cleaner, have you tried it?
Maximum graphics card power (W) -170, minimum system power requirements (W) -500. The PSU should be enough for everything. And the video card is under load and should be warming up: The maximum GPU temperature (in degrees Celsius) is 99. If 60-75 is present, this is quite normal.
This was not the case before?
PS The power supply unit is heating, but it should not be heated either, which power supply unit, how much is its temperature norm? Oh yes, and can we also say the position of the PSU - top / bottom?

I cleaned everything from dust when I put it on a new mother with a process, there were no jambs on the old system, but I don't know, maybe the vidyuhu limited the percentage there: / but even if so, it shouldn't be like that anyway

nicebloodx
I've updated the questions above.
Duck is a new mother with a processor, of course, the heat dissipation of the processor has increased, compared to the old one, the temperature of the outgoing air has actually increased, and if the power supply unit is located on top, then it is clear why it has become hotter (hot air passes through it).

nicebloodx
Here is a simple example of air movement in the case, study and draw conclusions (hot air is lighter than cold air, because it tends upward):
Photo

VOVAN WOLF Are
you saying that this is the norm and it should be so? because you can warm your hands, I'm afraid that the whole thing will burn out, although 650w should be enough

in about 20 minutes I'll tell you which company bp, and yes, the position is top

nicebloodx
Once again:
1) BP location - top / bottom;
2) What processor did you switch to?
3) The more demanding the game, the more it heats it up, for example Stellaris does not heat the video card more than 65, while ARK heats up to 72 easily, but this is an example of a modern video card.
4) Ventilation of the case, I repeat, study the circuit, try to find an additional heat source, it can be a new processor, hot air tends to the top, how heat is removed.
5) The temperature limit for the power supply unit, according to the characteristics, what is the norm for it, it is stupid to determine by the hand, do you need specific numbers to draw conclusions? In theory, an excessively high temperature leads, if I remember correctly, to swelling of capacitors, and eventually failure.

VOVAN WOLF
I wrote that I switched to i5 2300, thermaltake tr2 rx 650w power supply

nicebloodx Looked
through somehow. AMD has a calculated thermal power (TDP) of 65 W, Intel has 95 W. Intel is hotter, here's an additional 30 watts of heat. The top location of the power supply unit in system units is becoming more and more unacceptable, since the power supply unit throws hot air out, all from the video card, processor, RAM, motherboard. and got a couple of degrees.
There are two ways to control temperature: qualitative and quantitative. Quality - lowering the air temperature. Quantitative - the change in the volume / speed of the passing air stream. These methods can be combined. If you want to reduce, dilute with cold, i.e. additional fans, or change the case, put the power supply unit down, here you can think of a lot with the old case.

VOVAN WOLF and attach additional fans to the power supply? Well, I mean to connect them, this is again the load on the power supply :) and how is it quantitative? I can't understand, does it mean that the parts (generating heat) were located away from all this?

nicebloodx
Quantitative - increase the amount of air passing through the hull. Here's an example: there is a complaint about video cards with a turbine, they are noisy, why? The answer is simple - one turbine draws out the cooling of the video card, while a video card with two / three fans operates more efficiently, has low temperatures and noise. So the turbine must have higher rpm and drive out a larger volume of air to compete with two / three fans of an unreferenced video card. The same principle applies to you, in order to lower the temperature, if it does not suit you, increase the ventilation volume of the case, i.e. put additional fans on the case, unless of course there are none.
Look at the photo again:
Spoiler
What do you see there? And you see a simple picture of how hot air is expelled through the power supply unit, but it turns out that the higher the temperature of the video card, processor, even the HDD, the higher the power supply temperature. How to lower the temperature of the PSU? Consider the current photo that I added, there are the following options:
1) Put a fan on the back wall, it will dilute the air flow in front of the PSU fan, thereby lowering the temperature.
2) Perhaps opposite the processor, in the side wall, there are holes for a fan, you can add it by screwing it on, this will also dilute the air.
3) The roughest option is to replace the case, or alter the current one, in order to install the power supply unit from below, if space allows and the case is not a pity. Drilled holes for the PSU fan at the bottom, come up with fasteners, and as a result you will get air intake from the bottom of the cold one, i.e. the ventilated flow through the power supply unit will become independent, separate from the entire system unit, and you will get a hole in the top.
4) Perhaps, if there are other places for fans, and again, increase its volume, perhaps there is an SLI in the side wall of the fan mount.
5) In general, just score and put up with it, maybe this is the norm for him.
Fans do not significantly increase power consumption, by some 5-10 W. They are connected to the mat. board or power supply unit, this is what an adapter is, or whatever you want, through the mat. the board, their revolutions are regulated (although not everyone has such a setting), through a power supply unit only 100%, or a 7-volt adapter. By actually adding fans, you not only increase the volume of air passed through, but also the noise and dustiness of the fans themselves, but also of the case / accessories.
All my ideas, I didn't say anything new, maybe I made a mistake somewhere, others will correct / suggest, but so far somehow. Come up, craftsmen, turn on thinking, ingenuity.

nicebloodx post a
photo of the system unit with an open lid and a photo of connecting the power supply to the mother ...

corrected the situation, pulled the system unit out from under the table (so that there was more ventilation), in addition to this, I removed the power supply unit and put it on the system unit (the cooler up naturally). the power supply does not heat up, only the percentage itself after turning off the toys (overwatch, wow) with temperature flashes, jumps 40-50, mostly 40-45, as if it loads something, because when turned on there are about 37-38 degrees percent and not increases if idle is

nicebloodx
So it is, you run it on a cold one, heat accumulates in the gland, as soon as it has leveled off, it will keep it that way. Note, after all, distributions / browser / background programs, everything that uses the processor, after all, albeit slightly loads (you run programs). And it can also be noted that after a few hours of operation, it becomes warmer in the room, since the PSU + Video Card + Processor generate heat, also as a factor. In general, the cooler itself is probably a standard, unpretentious one. In general, there is solder under the lid, if you get a good cooler (the processor has a thermal power of 95 W), then the temperatures will be lower. In general, his maximum is 72.5 degrees, this is normal. In idle time, you can achieve temperatures and lower if you replace the cooler / thermal paste with something serious. The temperature is 40-50, respectively, this is its cooling, accumulated heat, after playing at high temperatures, it is released into the environment, and bounces back at 40-45. Once again, the percentage is hot, 95 W is too much for such a processor today.

VOVAN WOLF
understood, thanks, the main thing with the block is now figured out, otherwise I thought everything was dead, hence the overheating