[Alcaudon]

Quote:

 
Are you going to use MOSFETs or BJTs?Most of what I have read in various forums indicates "air wiring" MOSFETs is not necessarily a good idea. I am not sure about BJTs.

 
Mosfets. Yes, I've read the same, this is what holds me back. Guess that I'll have to stick to a "conventional" building and forget about low profile and tubes sticking out

 

 

[jdkJake]

 
Mosfets. Yes, I've read the same, this is what holds me back. Guess that I'll have to stick to a "conventional" building and forget about low profile and tubes sticking out

 

 


You could always bottom mount them from the underside of the board and use a bottom mounted heatsink. I have seen that done before. Not sure how well it conducts heat, but, it is an option.

 

[Alcaudon]

Quote:

 

You could always bottom mount them from the underside of the board and use a bottom mounted heatsink. I have seen that done before. Not sure how well it conducts heat, but, it is an option.

Yes, is another option, but I'm a bit concerned about heat accumultaing on the board due to air convection. Also, it won't reduce the height of the case.

 

[Sathimas]

wouldn't it be possible to reduce the bias-current for less heat dissipation?
 
I mean 100mA is way enough - I have a JISBOS buffer here biased to 20mA and this
is more than enough to make my DT880 sing I hear no difference to my M³.

 

[dBel84]

FWIW my original EHHA had airwired mosfets and beefy off board heatsinks. I have close-ups somewhere but this will give you an idea. 
 

 
The front and rear plates were 10mm copper, the entire thing got to about 40 odd celsius, but that was crazy biasing ..dB

 

[jdkJake]

Yes, is another option, but I'm a bit concerned about heat accumultaing on the board due to air convection. Also, it won't reduce the height of the case.

 


No, I meant to take a page from the playbook of a master:

http://www.head-fi.org/forum/thread/85561/post-pics-of-your-builds/3225#post_4153931

(see Ferrari's kumisa III build on this page)

 

[BK_856er]

 
Quote:

 
Those look nice BK. The new versions look to be far simpler and certainly less expensive to build. I am really glad to see Ti fixed the mounting hole locations. They should be far easier to work with as well.Do you still need to heatsink the transistors? Or is that now optional?

My understanding is that the transistor heatsinks are no longer required, but that conclusion is only based on the fact that Ti does not mention them on his v2 pages.  I should probably get confirmation on that detail.
 
In addition to the changes in the circuit, there is a hefty ground layer covering most of both sides.  Definitely took more heat to solder!
 
BK
 

 

[jdkJake]

FWIW my original EHHA had airwired mosfets and beefy off board heatsinks. I have close-ups somewhere but this will give you an idea. 


 





 


The front and rear plates were 10mm copper, the entire thing got to about 40 odd celsius, but that was crazy biasing ..dB

 


Wow, that is crazy small. Very nice.

What gauge wire did you use to airwire the MOSFETs? Did you make any attempt to match the lengths (either in physical length or electrical characteristic)?

 

[dBel84]

I kept the lengths the same but nothing else was considered. iirc the gauge was 20 ..dB

 

[holland]

I would keep the gate resistors on the pin.  Other than that you should be ok.

 

[jdkJake]

Quote:

I would keep the gate resistors on the pin.  Other than that you should be ok.

Interesting. Why is that? To keep the signal level higher until it actually reaches the gate? Is this for better noise rejection?

 

[holland]

Quote:

Interesting. Why is that? To keep the signal level higher until it actually reaches the gate? Is this for better noise rejection?

It's to help minimize/suppress oscillation at the gate.  MOSFETs have both parasitic capacitance and inductance that work together to form a tank/reservoir that can be excited into oscillation.  The R in series is to slow the rise of the gate voltage (RC filter, with C from the gate capacitance), and hence slow the switching speed of the MOSFET.
 
In short, stick it on the pin, as close as possible.  The EHHA layout has this, so don't lose it.

 

[jdkJake]

Quote:

It's to help minimize/suppress oscillation at the gate.  MOSFETs have both parasitic capacitance and inductance that work together to form a tank/reservoir that can be excited into oscillation.  The R in series is to slow the rise of the gate voltage (RC filter, with C from the gate capacitance), and hence slow the switching speed of the MOSFET.
 
In short, stick it on the pin, as close as possible.  The EHHA layout has this, so don't lose it.

Thank you! 
 
I learned something tonight. Makes it a good night indeed!
 

 

[BK_856er]

Officially, another one lives!
 
Took me a couple tries to get the grounding right, but it's playing music and it sounds mighty fine. 

 
Critical listening will come later with a proper setup, but I can already tell that this amp has some very nice qualities and the two of us are going to get along just fine.
 
BK
 

 
 
 

 

[BloodInside]

 
Quote:

So, BloodInside, are you listening yet?

No, unfortunately not. I didn't get any chance to work on it during the weekend. I should receive my E12 replacement parts this week, so I'm probably going to start working on that first. Based on 

[*] jdkJake concerns I also decided to snap of the power supply sections from the amplifier so I can move the amplifier away from the transformers.

[*]  

[*] I'm quite excited to start listening. My Little Dot MK IV is a bit underpowered for driving the Hifiman HE5-LE (but still sounding great), so I'm curious what the EHHA will do...