With the fan on my 350W antec before she blows?

The scientific answer is probably along the lines of measure the heatsink temps. I am no electronics engineer but believe the psu components to be a lot more robust than CPUs, so heatsink temps up to say 70C should be OK (to cause an argument).

The Heath Robinson answer is, "what, you still have a fan in your psu?" I don't, but then I do have 'positive' case pressure that certainly generates flow through my psu. Random positioned thermistor on my psu heatsink indicates max heatsink temp about 69C, air exhausting my psu is about 45C. It has lasted like this about 4 weeks so far

So go on, take it out. See if your psu enjoys a sauna as much as you do

I love sauna

So you have put a thermistor in the psu? I guess that would require buying an electronic one, and I'm a bit weary about that. I could take a normal one and place it in the powersupply for a while.

By positive pressure, you mean you have more air input then output? I have a 120mm fan sucking air into the fan, and two 80mm ones blowing it out. I'm currently planning on removing one of the 80mm ones, as soon as I get tools to cut the grill out from in front of it.

The hard part is, that currently I have the same knob on my controller controlling the 2 80mm case fans at the back, and the fan on the psu. It makes it kind of hard to figure out what is producing the noise. I guess I could allways unplug the two case fans to see... For some strange reason, the fan in the powersupply is making a very strong noise. I get the feeling that it might not be especially good quality. The problem is, that we don't have an especially good selection of silent fans here. Most people are buying papst, but I feel they don't stay silent for very long.

I noticed somebody mentioned having a tunnel from the powersupply, with the fan at the end. Does it make a big differance? Would that be similar to the 'shroud' I have on my watercooling?

Yes, but nothing fancy. I just converted one of those cheap indoor/outdoor digital thermometers so that the indoor thermistor is also on the end of a cable (desolder it from the pcb etc). Then when I get worried about my temps I have 2 'mobile' sensors I can place anywhere I want.
Again yes - my only case fan is a 120mm front intake. That air has to exhaust somewhere - and as I have a cheapo psu (which is therefore relatively low density component population) the airflow through the psu is OK and a fair proportion of my case air does exit through the psu.

ps - you may want to drop Dorothyan email - I dont think postage to Finland would be too bad as we are all EU. Gives you a choice of Panaflo and NMB for 80mm as well.

What kind of case temps do you get like that? I could possibly do a deal with somebody from england, the problem is my lack of a credit card. (and I don't want one...)

OK, just put my 'indoor/outdoor' with 1 sensor by the 120mm intake, 1 by the rear case vents (there are stamped vents for 2 * 80mm which are unpopulated) by the CPU which I then moved to the psu exhaust vent.
22C (room temp on wall thermometer)
20.5C (thermistor by 120mm intake)
26C (MBM report of mobo temp)
28C (DTemp report of my 'cuda SMART temp)
43C (MBM report of thermal diode - I am folding 24/7)
25.3C (thermistor by rear case vent)
41.9C (thermistor by psu exhaust)
No problem - I am happy for you to post me € notes if you want - just PM me if you want to go this way.

I live in Finland and have dealt with Dorothy. You can wire the money straight into her account. IIRC my bank (OKO) took a 6-7€ fee for this procedure, and it took 3-4 days for the money to show up in her account. The fans were at my doorstep five or so days after that. Dorothy was very nice to deal with.

BTW: very nice temps with that setup, dukla.

If I decide to modify my powersupply, and make a fan either blow from the behind, or through the bottom, what else then airflow should I be looking at? The thing is, my case is dampend, and everything producing sound inside the case is silent. The powersupplys fan is obviously almost outside the case, and the noise is more high pitched then anything else coming out.

As long as the air is flowing over the powersupplys heatsinks, am I on fairly safe waters?

Well, as soon as you open the psu you void the warranty and you are on your own. After that without a fairly decent lab, especially to measure temperatures in and around the psu, it all gets 'experimental' and depends how much risk you are prepared to take.

A 'simple' mod that moves the existing 80mm fan on a psu from the back (AC input side) to the front (DC exit side) preserves the manufacturer's cfm calculations, but allows you to damp the fan mountings and moves the fan 14cm further inside the case.

After that it gets more 'unknown': on the one hand the psu manufacturer designed (may be an excessive assumption for some psu ) the psu to cope with ambient temps up to 40 or 50C, as well as to run at peak loads for extended periods. They probably also engineered some factor to allow for the situation where the psu fan is the ONLY case ventilation (correlated effectively to the ambient temp assumption). Last, they may even have deliberately 'designed' a stamped grill for EMI reasons, and engineered the fan speeds with the stamped grill as a constraint. So IF this is all true, and your environment is different (say ambient 25C, normal DC load 90W, and you swap a stamped grill for a wire finger guard) then there should be room to downrate the fan speed.

Looking at Mike's latest review of the 400W Nexus, the fan there is rated (approx as not sure precise model) 41cfm @ 12V, and runs at 7.1V at 90W (DC) load, so extrapolating that 24cfm should be reasonable for the average 90W load, which is a Panaflo L1A at 12V. As per my para above, if you are prepared to take more risk (or can measure all the psu temps, or assume other things the psu designer couldn't know for your environment like you have decent case ventilation) then dropping that L1A to 5V is an option. (or somewhere in between 12 and 5).

But apart from the standard potential electrical/electronic safety hazards working on a psu, airflow/cooling is your main concern. It cant be overlooked though that if you do remove a stamped grill you change the EMI characteristics, if you break it (up to and including setting fire to your house ) you own both parts (and some discussions have formed an opinion that is how your house insurers may react as well).

Stupid idea number one:

What if say a person, me, just disconnected teh psu fan and mounted a panflo on the outside of the psu and cutout the grill - would this facilitate enough airflow to satisfy the psu without having to tear into the psu which is what I am DREADING TO DO and I do not right now have the $ to buy a good silent psu?

If you mean without removing the internal fan, that's a funny one. Never tried that one before, but for it to actually have a point, the internal fan would have to be disconnected, which means you have to ** go inside** With the inside fan turned off, there would probably be a fair amount of turbulence noise caused by the impedance of that fan at the intake of the outside fan. You would definitely lose some airflow too. Not good.

dukla2000 -- some impressive posts in this thread! Your 41.9C at PSU exhaust does not seem excessive, btw. Is that the highest temp there? I found that the temp of the exhaust varies a bit at different points of the exhaust hole.

Yes Mike that is what I meant hehe, I think I may just save up and buy a less noise generating psu the idea of tearing into mine and getting shocked to death just does not appeal to me. Unless I can build a nice carpet muffler for it.

It is unscientific - I just try balance the bead of the thermister in a gap in the wire finger guard, usually close to the centre of the guard. I have found if I loop the wire in and out the guard and the thermister is out of the case by about 0.5cm (or more) the temp drops as it is being affected by the room ambient, so I try keep the thermistor just inside the finger guard. I haven't tried any top/sides/bottom for any variation. I will try that at some stage - at the moment the UK is having its heatwave for the year and my room temps are up to 25 which is playing havoc with a minimally coolled setup, so the front Papst is cranked a little louder than I like

Occasionally I push the (thermister) wire right through the finger guard so the thermistor parks in a gap in the (DC regulator) psu heatsink, and that gets readings up to 48C.

frosty - another challenge is how would you mount the panaflo - the existing fan is using the best set of holes. And as Mike pointed out, having the obstruction of the old fan right by the intake of your 'flo is not optimum: fan noise caused by an obstruction close to the fan intake is worse than the same obstruction the same distance from the fan exhaust.

Yo Dukla after going home at lunch I see what you mean, I am just chicken to open up my psu cos of all the articles I read on shocking myself to death! :} But I am working on doing it anyways.