Newbie building JFET Zen v8

[summilux]

Hi everyone,

I've been building tube amps for a few years and I thought I would try something different for a change. I will try to build the Zen JFET with a capacitance mutliplier PS ( IRF240). I've been going through as many of the first watt and JFET posts as I find and I have a few questions:

Heat sinks: What temperature rise is permissible/optimum. Should I be aiming to keep that increase to 10oC or 20oC or some other value? How good are the numbers on the Conrad and Thermaflo websites to guide me in heatsink sizing? How much will anodizing affect the real world performance of the heatsink?

Transformer. With the voltage loss in the PS, I will need a 40V transformer to get more or less 50V into the amp. I am calculating about 105W total power dissapation in one channel, 15W in the PS, 40W across each of the 22 Ohm power resistors (I don't want to use the lightbulbs) and about 10W in Q2. One post I read suggested that I size the TX to 4X the idle, bringing me to 420 VA. Is this appropriate?

Thanks. I'll probably have more questions as I go along.

Jonathan

[Alain Dupont]

summilux,

>> Heat sinks: What temperature rise is permissible/optimum.
50 to 53 degree Celsius max
Nelson says that: if you can keep your hand's on them for about 3 to 4 seconds
Better use a meter... Consider +25/30 degrees over current temp.

>> How good are the numbers on the Conrad and Thermaflo websites to guide
>> me in heatsink sizing?
The numbers are correct. Conrad are easier to work.
Take attention to "normal convection" mode, otherwise you will need a fan.
And also depends on pads, mica and grease, silpads ...

>> How much will anodizing affect the real world performance of the heatsink?
Black anodizing will radiate a little more heat. {just a few %}

>> 420 VA. Is this appropriate?
Average is 2 to 3 times the power dissapation ; You'll need 2 of them.
Yes, good choice, be aware that >100 watts by channel gives BIG heatsinks...
100 watts in pure Class A is huge brute power!

Where do you want to use this amp?

Regards.

PS: I am a A75 avid user!!!

Alain.

[summilux]

Thanks for the input Alain. I have a pair of Fostex 208 that I'd try it in. These are currently driven by a 300B SET with plenty of power fo 5W should be good enough fro these babies.

Jonathan

[Paul Ebert]

I'm in the same boat (a newbie wanting to build a JFet amp). I've been looking for heatsinks and the only one I've found that looks like it can work is this one:

http://rocky.digikey.com/scripts/Pro...45&M=392-180AB

If you find others, I'd be very interested.

Paul

[Paul Ebert]

Oh, I forgot to mention. The heatsink in the link dissipates 260 watts and costs $120 according to digikey (or it may cost $112, depending on which page on the website you look at).

Not too bad, but I'd like to find somthing with a better form factor.

Something cheaper (say, surplus) would be nice as well.

Paul

[summilux]

Paul,

I found this a Conrad Heatsinks:

http://www.conradheatsinks.com/produ...le_f.html#MF30

It's 52 AUD + shipping and should be able to dissapate 100W with a 25 C rise in temp.

Jonathan

[Paul Ebert]

Thanks, Jonathan. That does look good.

Have you given thought to the circuit board? I'm hoping that someone will do some sort of group buy in the future or at least post a layout. I'm not sure I want to tackle laying one out and having it built on my own. On the other hand, I do not know if point to point is appropriate.

I'm also thinking I may wait for one of the future designs that improve on the performance.

Paul

[Paul Ebert]

I took a deeper look at the sink available from Digikey (a Wakefield, btw) and it will not work. It looks like it will rise over 40 degrees C dissapating 100 watts.

[hifimaker]

Hi Paul,

Conrad Heatsinks shipped my 4 of their largest heatsinks. The shipping cost by Aussie Post was only $15! Multiply their c/w ratings by a correction factor of 1.4 to correct for normal room ambient temperatures.

The quality is awesome. They are also powder coated rather than anodized, which I personally prefer.

-David

[Paul Ebert]

A correction factor of 1.4? Well, that makes the MF30-2F that Jonathan mentioned borderline IMO (rise of 35 degrees C). The MF35 goes up 29 degrees C, which is probably OK, but it doesn't have the nice flanges.

I am glad to hear that the shipping isn't too bad, though it probably is more to my neck of the woods.

I'm wondering if subsequent versions with ccs will need less dissipation. Or, will they need more?

Paul