Fan....Push or Pull?

[Deric]

It's weird. Every amp I have has some hum (vibration) from the PT. On most of them you have to put your ear against the cab to hear it. This one's not noisy at all till bolted into the cab....

I did just spend a few minutes messing with it and if the chassis is bolted in slightly crooked it's barely noticeable. Guess that'll work for now.

[Alexo]

I vote for push-pull!

[martin manning]

I double checked the direct. I powered on 3 SVT's if i put a piece of paper against the grill it sucks the paper and holds it. so the air flow is outside air - into the amp.

Not enough words here... needs a picture! The attached drawing shows what I was thinking re recirculation, and the clipped photo of an SVT shows an arrangement very similar to Deric's. Note it appears to have the fan exhausting directly out of the box. I looked at another SVT model in the local GC yesterday, and powered it up to check the airflow. It had the fan mounted against the inside of the front grille, exhausting out through the grille cloth. That set-up is better in that the heated air exhaust is separated from the inlet, which was the open back of the cab (with a perforated metal cover, of course). Pulling heated air back in is like positive feedback- not a good thing.

[Structo]

I guess my argument against having the fan blowing in on the amp is that it can create hot and cold spots, especially when the power tubes are involved.
If the air stream is focussed on only a couple of a quad of power tubes then the cooling will be uneven and I would think it could affect the operation of those tubes after a while.

Perhaps in the long run it doesn't hurt anything but that was something I read about why it isn't a good idea to have the fan blowing on the tubes.

[vibratoking]

Cold is just an absence of energy. You want to get the energy out before it has the opportunity to transfer to other materials. Push or pull could be the best depending on the situation, as Bob-I mentioned.

Creating positive pressure inside the box is great if you are going to control the flow with baffles or in some planned way. If not, the effectiveness is lost. You have to have enough in vs out flow for the fan to work optimally. If you starve it on one side or the other, you diminish the effect.

[Bob-I]

Cold is just an absence of energy. You want to get the energy out before it has the opportunity to transfer to other materials. Push or pull could be the best depending on the situation, as Bob-I mentioned.

Creating positive pressure inside the box is great if you are going to control the flow with baffles or in some planned way. If not, the effectiveness is lost. You have to have enough in vs out flow for the fan to work optimally. If you starve it on one side or the other, you diminish the effect.

Positive or negative will have a similar effect. The key is air movement so careful planning is important. Just blowing or sucking won't take care of business, the air must flow. I've seen datacenters where one rack pushed hot air on the next rack then the folks are puzzled why one rack overheats and the other doesn't.

Come on folks, I thought there were some engineers here? Thermal engineering anyone?

[Alexo]

Honestly, I think we're over-thinking this a bit.

As long as there is a modicum of air circulation, things should be fine in a tube amp. Most components are rated for near-boiling point operation, and lord knows the kinds of environments tubes were routinely subjected to back in the days before they became so boutiquey.

Having said that, the cathodes do need to be heated to emit electrons, so don't point a big honking computer fan straight at them. In the case of the OP, I would put it on 'exhaust mode,' as that looks like it would have the most even cooling effect across the power and rectifier tubes.

That fan may be too powerful, looks like the same type I used in my cat's litterbox to push the air out into a dryer hose hooked up to the chimney. Strong enough to create a draft on the other side of the house.

...the answer was obvious in that case, re push/pull, btw

...and I know we want to keep things cool, but dollars to doughnuts, those tubes will sound better HOT.

[dynaman]

Honestly, I think we're over-thinking this a bit.

As long as there is a modicum of air circulation, things should be fine in a tube amp. Most components are rated for near-boiling point operation, and lord knows the kinds of environments tubes were routinely subjected to back in the days before they became so boutiquey.

Having said that, the cathodes do need to be heated to emit electrons, so don't point a big honking computer fan straight at them. In the case of the OP, I would put it on 'exhaust mode,' as that looks like it would have the most even cooling effect across the power and rectifier tubes.

That fan may be too powerful, looks like the same type I used in my cat's litterbox to push the air out into a dryer hose hooked up to the chimney. Strong enough to create a draft on the other side of the house.

...the answer was obvious in that case, re push/pull, btw

...and I know we want to keep things cool, but dollars to doughnuts, those tubes will sound better HOT.

Yeah, what he said.

I've read that it's not such a good idea to force cool air directly on power tubes.

When I first started stuffing circuits into toolboxes, I used a fan and located vents so that fresh air was drawn across the majority of the hot components. Nothing crazy - just a bit of circulation. These amps ran ice cold. Over time, I realized that simply venting the box tops was more than adequate and I quit using fans.

It really doesn't take much circulation to keep things cool. On occasion, I've also used holes in both the top AND bottom of an amp to take advantage of the "chimney effect", but most amps won't allow that to work well because of their chassis' being nearly sealed against a cab.

IMO, if one has to force a bunch of cold air onto a component to keep it running properly, there's prolly a design problem.

[PCollen]

Another reason for push is that if the area is dusty, and the fan is blowing out, air will be drawn in through the other gaps in the cab -- meaning dust will build up in all the other gaps, rather than being drawn through a fan that can be filtered if need be.

..and ANOTHER reason is that if you try and pull the hot air out, you need a closed plenum in which to mount the fan, not in the middle of the rear screen grill which is a large open area. In the latter case, you are creating a vacuum behind the fan and cool air from 'outside' will just flow right around and back out through the fan blades..VERY inefficient. By pushing cool air in, you are dispersing the heat in the immediate vicinity with cooler air, albeit not as cool as if ther were a duct leading to an entry point further from the tubes. AND, you should not blow cool air directly on the tubes as that creates a heat imbalance from one side of the tube to the other. Probably the best thing to do would be simply to put a grill in the top of the cabinet ABOVE the tubes and just let the heat naturally rise upwards and out of the cabinet..that's all you really need.

[Structo]

Yes but he is trying to keep the mosfet in his VVR cool so not the tubes per say but the area where the mosfet is mounted.

[PCollen]

Yes but he is trying to keep the mosfet in his VVR cool so not the tubes per say but the area where the mosfet is mounted.

In THAT case, take the example of your PC CPU chip with a small, dedicated fan blowing directly on it.

[rdjones]

Come on folks, I thought there were some engineers here? Thermal engineering anyone?

Well, nothing's been suggested about pressurized air being denser and therefore being able to transfer more thermal energy. (?)

Maybe the FET just needs a bigger heatsink.

rd

[rdjones]

I lent it to a buddy for a while and he managed to blow 2 FET's Not sure what happened as I've never had an issue and I'm pretty sure I play it a lot harder than he did.

I suspect the FET is working harder when more "attenuation" is being used. The harder you play it the less work the FET circuit is doing.

rd

[M Fowler]

You have a push-pull amp your going to need push-pull fans, simple

[greenbottle]

These fans are designed to push cold air. Causing them to suck hot air
seriously shortens the life span.

Why does he need a fan. Because he is blowing transistors.

Does the transistor have heat sink. Is the transistor making good themal
contact. Is the heat sink adequate. Is the heatsink dissapating the heat from the transistor or is it absorbing heat from the rest of the amp.

Can you relocate the heatsink. If not can you shield the heatsink from extraneous heat sources.

If all or any of these options are non viable then pull the device out and build a lower wattage amp.

Alan