How to build the F5

Jump to Latest
#342
Bksabath said:
WELL WELL WELL
We got in the house some one that at one point suggested to solder the mosfets to the cooper directley should I fetch that post...


Any way one thing I would like to make clear the main reason why I am using the heat spreaders is so I will not strip the treads on the heat sink, this is just a Mule so I will keep triing different things, different Mosfet different number of mosfets balanced and what not may even turn the thing in to the A75.
Click to expand...

And I would still do it the same way, what's wrong with this approach?

BTW, Denon used heat spreaders in their flagship POA-S1 amp.
 
#345
Slow fans

Amen, 6L6, on the "slow fan" concept.

I'm using two 120VAC shaded-pole, 90mm fans in my F5 version. I'm doing so, because the two Conrad heatsinks will effectively be INTERNAL to the amp chassis. (I'm going to have wood sides on the left and right of the chassis, to match my other audio equpment).

As such, I have two rear-mounted fans, that will "pressurize" the internal of the chassis (PSU, PCBs, capacitor banks, VU meters, etc). I've milled a 3/8" relief along the entire bottom of each Conrad heatsink plate, to permit air to flow from the internal of the chassis, under the heatsink, and then upward between the heatsink fins and the (final) wooden sides. The air will exit at the top of the chassis.

I'm using a dropping resistor (about 560 ohms, if I recall), to bring down the blade speed on both fans. They are barely audible on the workbench, and should be "silent" in the chassis when it's all buttoned up. You're right--it doesn't take much air movement to be a LOT more effective than convection cooling, especially with the surface area of the Conrad heatsinks....!

If I wasn't going to enclose the heatsinks, the use of fans probalby becomes moot; but since I'm going in the direction I am, I see fans as a necessity.

Ken
 
#347
AndrewT said:
Peter,
think about it from the other direction.
Take a loaf of bread, a lump of stone and a block of aluminium outside and leave them to cool overnight.


Andrew--are you suggesting I use a loaf of bread as a heatsink on my next F5 variant? Would you recommend Russian Rye, or Whole Wheat?

Hey--take it easy on me..... 😀

I'm just a "newbie" trying to learn about MOSFETs, thermodynamics, enthalpy, entropy, eigenvalue transfers, bread, and such..... 🙂
Click to expand...
 
#352
Peter Daniel said:
I don't use any special machinery, this is all I need to work with aluminum. I recently added sandblaster to my workshop.
Click to expand...

Peter--thanks..... you do great metalcraft....!

I recongize a table saw in the middle of the picture (and I'm assuming with a metal-cutting blade).

To the left--is that a jointer-planer, or maybe a large routing table?

P.S. Have fun with the sandblaster. If you experiment with different blasting media (walnut shells, glass beads, etc) you can get some very interesting finish effects on aluminum, copper, and other soft metals......
 
#353
Bksabath: Re: MIC502 fan managment IC
quite handy got a few interesting functions.
Such as set point at which it can be turned off.
So no fan while the amplifier warms up for faster setling times

I checked out the data sheet on the MIC502. Be careful about your PW modulation frequency. The data sheet indicates it might fall into the audible range (30-1000 Hz), depending on the choice of components. Might be another source of hum to contend with.....
 
#354
As such, I have two rear-mounted fans, that will "pressurize" the internal of the chassis
Click to expand...

I am a relative noob here, however I have about thirty years of experience servicing radar systems and lately, accelerators used for radiation therapy.
Many years ago when I worked for a large defense contractor (Raytheon), I had an interesting discussion with a design engineer regarding air flow and cooling. The details of the conversation are a bit sketchy, however I do remember the engineer indicated in most cases pulling the air through a chassis is a more effective cooling strategy than forcing air into it.
As anecdotal evidence, most assemblies I have serviced also draw air through the unit instead of force feeding it.
 
#356
AndrewT said:
The copper feels hotter because it is transferring more heat to your fingers.
The aluminium feels cooler because it cannot conduct heat as well.

But there is a part you have overlooked.
The Specific Heat of the two metals is different.
Even though they are both heated to the same starting temperature, they do not contain the same energy.
The volumetric heat capacity are 3.45:2.42 for copper:aluminium.

The copper started off with ~42% more energy. If they did cool down at the same rate in air, then that confirms that the heat flow out of the copper was ~ 42% more than the aluminium.

Copper will always make a better heatsink than aluminium when you compare on a dimensional basis (volume and/or area)
Click to expand...

This makes sense. The Cu holds more energy than the Al (if both are the same size and start at the same temp). For Peter Daniels experiment showing equivalent decreases in temp, that would mean that the Cu bar is actually releasing more energy into the ambient air over the same surface area. I guess this is why most CPU heatsinks are copper? Better efficiency for smaller spaces?
 
#357
6L6 said:
Ok, let's look at the schematic. Remember that I am wiring it for 120v operation, so the transformer primaries are in parallel.

........One red primary is connected to AC hot through a thermistor.

........one of which is connected to the AC through a thermistor.

If I am doing something wrong, somebody please tell me! 🙂
Click to expand...

Jim--I'm certain you've thought of this, but here goes--make certain (use an ohmmeter) that you have "paired" the transformer primaries correctly. I.e., one black/red primary winding goes thru one thermistor--and the OTHER black/red primary winding goes thru the OTHER thermistor. If you are not careful, you can end up with one black/red primary winding going thru TWO thermistors, and the other black/red primary winding not going thru EITHER thermistor. :headbash: (If you end up doing this, please invite me for the first time you fire up the power supply--and "fire up" may be the accurate term!)

Ken
 
#358
sorry but i think the last four pages need to be moved over to the f5 thread . 6L6 started this thread as a tutorial to building a box stock f5 and you guys have just turned it into another egofest totally unrelated to the original premise .

WTF does Bksabath's build have to do with building a stock F5?

lets get back on topic please .
 
#360
Retraction of useful build data.....

6L6..... Jim.....

Per Woody's comments, please disregard my cautionary comments about "pairing" the primary windings on your F5 power transformer to avert a disaster upon start-up. I had provided this info, per your comment "let me know if I am doing this right".

Same goes, for my comment about potential use of slow fans in the F5 build. I had responded to your query on the same subject, but I suspect it might have been an unwelcome response (by some).

Are you close to full-up testing? I set my biases this weekend.

(Parenthetically, it appears some others have already established their biases)
 
You must log in or register to reply here.
Top Bottom