Scaleable zen case
I have decided to build my zen 4 using the heatsinks and the 10 kilo EI transformer shown at the top of this thread.
Here is quick drawing of my idea of cooling:
Since the sinks are 220 mm long the rear of the 3 or 5 mm case sides (blue) will not have any sinks.
The sinks will be bolted to the sides, maybe using thermal grease between, and the FETS will be mounted on the sides, thus hoping for cooling through the blue sides over to the sinks...?
Would this be possible or do I have to mount the FETS directly on the sinks?
My case here would be scalable, and I can add more small sinks to the rear sides, if things get to hot, couldn't I?
If this get too warmy, I also have the option to lower BIAS.
Front will be oak wood.
I have decided to build my zen 4 using the heatsinks and the 10 kilo EI transformer shown at the top of this thread.
Here is quick drawing of my idea of cooling:
An externally hosted image should be here but it was not working when we last tested it.
Since the sinks are 220 mm long the rear of the 3 or 5 mm case sides (blue) will not have any sinks.
The sinks will be bolted to the sides, maybe using thermal grease between, and the FETS will be mounted on the sides, thus hoping for cooling through the blue sides over to the sinks...?
Would this be possible or do I have to mount the FETS directly on the sinks?
My case here would be scalable, and I can add more small sinks to the rear sides, if things get to hot, couldn't I?
If this get too warmy, I also have the option to lower BIAS.
Front will be oak wood.
Fet mounting
I would not mount the FETs in the way you have pictured here. If the sides of the case are aluminum, it might work, but even then there will be additional thermal resistance. Using thermal compound will help, but given that your sinks are already on the small side, I think this mounting arrangement is asking for trouble. If the case is steel, it will be even worse.
You should mount the FETs directly on the heatsinks, perhaps by cutting holes in the sides of the case.
-Dan
I would not mount the FETs in the way you have pictured here. If the sides of the case are aluminum, it might work, but even then there will be additional thermal resistance. Using thermal compound will help, but given that your sinks are already on the small side, I think this mounting arrangement is asking for trouble. If the case is steel, it will be even worse.
You should mount the FETs directly on the heatsinks, perhaps by cutting holes in the sides of the case.
-Dan
Thanks Sparhawk
I will mount them directly.
Caps
Can I have someones opinion on this issue:
If I decide to use two beer cans only the total capacitance will be 33000uF per channel. If I use all four of them on the first pic it will be 60000uF per channel.
My trafo is 800VA. Nelson recommends 25000uF per channel as minimum (?)
My case will be quite huge if I use four. Easier to build with only two of 'em.
1. Not a good thing to place the caps on top off the trafo ?
2. Do I get a better sounding amplifier using 60000uF per channel ?
I will mount them directly.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Caps
Can I have someones opinion on this issue:
If I decide to use two beer cans only the total capacitance will be 33000uF per channel. If I use all four of them on the first pic it will be 60000uF per channel.
My trafo is 800VA. Nelson recommends 25000uF per channel as minimum (?)
My case will be quite huge if I use four. Easier to build with only two of 'em.
1. Not a good thing to place the caps on top off the trafo ?
2. Do I get a better sounding amplifier using 60000uF per channel ?
1.
No idea if good or bad. But, my amp has four 22000uF caps arranged in a row about 1 inch above EI tansformer (300VA). I don't catch any problem.
2.
Well known is that the bigger the caps are, the smaller ripple and that the smaller the ripple is, the better. I however suspect the caps bigger than the amp designer's calculation would improve the sound.
Hoping this will help.
No idea if good or bad. But, my amp has four 22000uF caps arranged in a row about 1 inch above EI tansformer (300VA). I don't catch any problem.
2.
Well known is that the bigger the caps are, the smaller ripple and that the smaller the ripple is, the better. I however suspect the caps bigger than the amp designer's calculation would improve the sound.
Hoping this will help.
Here's a thought. If you don't mind having two boxes.
Put the transformer and one pair of caps in the first box, and the second pair and the amps and heatsinks in the second box. The first box can be ugly and hidden away, and it solves the possible problem of the transformer flux causing hum. I've also had a problem with the charging current peaks inducing hum.
You can add a small amount of resistance between the two pairs of caps (in addition to that of the wiring itself), to get a pi-filter.
Put the transformer and one pair of caps in the first box, and the second pair and the amps and heatsinks in the second box. The first box can be ugly and hidden away, and it solves the possible problem of the transformer flux causing hum. I've also had a problem with the charging current peaks inducing hum.
You can add a small amount of resistance between the two pairs of caps (in addition to that of the wiring itself), to get a pi-filter.
I got a quote from mmmetals for mm5060s: $166 for 6 pcs - I was planning to use 3 per channel. This was bit on a high side though not bad considering that they are new (also, I found out that they are .5C/W each - not bad at all).
Subsequently I picked up two 24" (huge!) sinks from MECI - one per channel. Just a note to anyone planning to buy these - they look anodized in the pictures shown but are not actually. I am planning to put both mono-blocks in one case with huge heatsinks on both sides - 24" tall tower. Just sent in a check to passlabs for PCBs!
Subsequently I picked up two 24" (huge!) sinks from MECI - one per channel. Just a note to anyone planning to buy these - they look anodized in the pictures shown but are not actually. I am planning to put both mono-blocks in one case with huge heatsinks on both sides - 24" tall tower. Just sent in a check to passlabs for PCBs!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.