Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Let's fix swt61's Beta22!
New Posts  All Forums:Forum Nav:

Let's fix swt61's Beta22! - Page 6

post #76 of 86
12/16/09 at 3:28pm
Quote:
Originally Posted by amb View Post
For β22, rather than adding any resistors, you could simply increase the resistance of R34 and R35. < snip >
Thank you for this... Pretty clear when you look at the schematic I see you spec 2W 5% metal-oxide resistors there as well.

I'll probably go that route as I prefer passive protection methods as opposed to having to take some action (pwr off, pull cans, etc).
Reply
post #77 of 86
12/16/09 at 3:35pm
Quote:
Originally Posted by amb View Post
For β22, rather than adding any resistors, you could simply increase the resistance of R34 and R35. This has the benefit of keeping the increased resistance within the global feedback loop, so the closed loop output impedance is not adversely affected. However, it will affect the maximum output voltage swing into low-Z loads. Hence, I would recommend this mod only for 32 ohm or higher headphone loads, not for speaker loads.
And this change can be made without any other issues? No other tweaks required?
Reply
post #78 of 86
12/16/09 at 3:52pm
Thread Starter 
Quote:
Originally Posted by Beefy View Post
And this change can be made without any other issues? No other tweaks required?
yes!
Reply
post #79 of 86
12/16/09 at 4:00pm
Quote:
Originally Posted by luvdunhill View Post
yes!
Groovy. I semi-recall discussions on a diamond buffer based amp (PPAv2?) where tweaking those resistors was all the rage, but the wrong value could tip the amp towards instability.

Though it was a LONG time ago that I was reading about that......
Reply
post #80 of 86
12/16/09 at 4:04pm
Thread Starter 
Quote:
Originally Posted by Beefy View Post
Groovy. I semi-recall discussions on a diamond buffer based amp (PPAv2?) where tweaking those resistors was all the rage, but the wrong value could tip the amp towards instability.

Though it was a LONG time ago that I was reading about that......
In general, that happens from going too low. In the case of paralleled devices where they aren't matched all that closely, current hogging can happen. Too high isn't as big of a concern, if you can handle the power dissipation, the potential for higher output impedance, and in extreme cases, less voltage swing. As Ti stated, none of these are really an issue here, given the caveats he gave.
Reply
post #81 of 86
12/16/09 at 4:09pm
Reply
post #82 of 86
12/16/09 at 4:12pm
Quote:
Originally Posted by Beefy View Post
And this change can be made without any other issues? No other tweaks required?
As long as you apply Ohm's law w/new resistor values WRT bias as Ti said:
Quote:
Originally Posted by amb View Post
Note that increasing the value of R34/R35 will impact your output stage bias adjustment. Ohm's Law says V = I * R, so if you want 120mA of quiescent current, and R is 1 ohm, then you need to adjust VR2 to get 120mV across R34 (and R35).
Reply
post #83 of 86
12/16/09 at 4:33pm
Thread Starter 
I don't think you have to worry about increasing power handing of those 2W resistors. In this case, these resistors were fine, it was the gate resistors that fried.
Reply
post #84 of 86
12/16/09 at 8:16pm
There is no need to increase the power rating of the output source resistors R34/R35. I would rather have these fry as opposed to the MOSFETs.

The gate resistor burns up as secondary "collateral" damage to a fried MOSFET. Normally, the MOSFET gate input is very high impedahce, and the gate resistor flows virtually no current. But when the MOSFET is destroyed, the gate usually becomes internally-shorted to the source, causing a large current to flow through the gate, and takes out the gate resistor too.
Reply
post #85 of 86
12/16/09 at 9:11pm
I picked up some to220 sockets from mouser and have yet to actually try them.

would it be wise, unwise or both (grin) to use sockets instead of soldering in the mosfets? I have the sockets ... I could use them ... but not sure if its good for things that will carry a lot of current.

otoh, it makes replacing the mosfets a 1 minute exercise, with sockets.
Reply
post #86 of 86
12/16/09 at 9:23pm
Thread Starter 
Quote:
Originally Posted by linuxworks View Post
I picked up some to220 sockets from mouser and have yet to actually try them.

would it be wise, unwise or both (grin) to use sockets instead of soldering in the mosfets? I have the sockets ... I could use them ... but not sure if its good for things that will carry a lot of current.

otoh, it makes replacing the mosfets a 1 minute exercise, with sockets.
not the best idea.

I bet they won't fit... unless you redrill the device higher on the heat sink, but then you'll make the heat sink not perform as well.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Let's fix swt61's Beta22!