Regulator Tweaks?
post-221642
Thread Starter
Post #1 of 10

thegopher

New Head-Fier
Joined
Nov 20, 2002
Messages
7
Reaction score
0
Joined
Nov 20, 2002
Posts
7
Likes
0
Anybody got any ideas to tweak 3 pin regulators to the max, other than those covered by data sheets?
 
     Share This Post       
post-221745
Post #2 of 10

Dreamslacker

500+ Head-Fier
Joined
Oct 7, 2002
Messages
555
Reaction score
10
Joined
Oct 7, 2002
Posts
555
Likes
10
1) polish the back
2) clean with alcohol
3) apply Artic Silver 3 thermal compound
4) mount a big heatsink
5) screw real tight to ensure maximal metal to metal contact area and squeezing out of excess compound
6) clean off excess compound

That should improve operating temperatures, leading to higher output current/ voltage (since temperatures are lower). Really overkill though. Especially the compound part.
 
     Share This Post       
post-221947
Post #3 of 10

ppl

Building amps and assuring water resistance.
Joined
Jun 21, 2001
Messages
1,772
Reaction score
13
Joined
Jun 21, 2001
Posts
1,772
Likes
13
put small 1-2 ohm resistors in series with the output and within the Feedback loop to improve stability. i have posted about this befor but can't remember where
 
     Share This Post       
post-222052
Post #4 of 10

thegopher

New Head-Fier
Joined
Nov 20, 2002
Messages
7
Reaction score
0
Joined
Nov 20, 2002
Posts
7
Likes
0
ppl-

Are there any limits on input and output capacitance for stability with this method and are LDO regs like the LT1085/1033 more susceptible to stability problems with high values of output C than the LM317/337 types?
 
     Share This Post       
post-222254
Post #5 of 10

ppl

Building amps and assuring water resistance.
Joined
Jun 21, 2001
Messages
1,772
Reaction score
13
Joined
Jun 21, 2001
Posts
1,772
Likes
13
The LT-1085 & LT-1033 are just an improved version of the LM series most all operating concerns are similar for Both. The LT reg's do offer slight improvements over the LM types However for most applications thay can be considered improved Replacements for the LM series. Also look at the Schematics on the data sheets and see the topology is also similar in fact about the same.

The reason for the resistor on the output is to allow the use of low ESR electrolytics and Film capacitors on the output. Input considerations remain the same
 
     Share This Post       
post-222297
Post #6 of 10

eric343

Member of the Trade: Audiogeek: The "E" in META42
Joined
Jun 23, 2001
Messages
6,033
Reaction score
11
Joined
Jun 23, 2001
Posts
6,033
Likes
11
Also, don't mount it using the hole! I've seen a bunch of people mount TO-220s like this (including Mr. Gilmore! [though his method largely eliminates the issues encountered]). It places pressure in the wrong place; the chip is in the plastic! Use a piece of aluminum stock or u-channel to press the /whole/ transistor against the heatsink.

When you use a screw through the hole, what happens is that the transistor can get slightly tilted towards the heatsink, "lifting" up the part of the tab where the chip is away from the heatsink, and reducing thermal efficiency.

http://sound.westhost.com/heatsinks.htm#flat%20packs

Oh, and paint that shiny anodized aluminum black using oil paint.
 
     Share This Post       
post-222576
Post #7 of 10

thegopher

New Head-Fier
Joined
Nov 20, 2002
Messages
7
Reaction score
0
Joined
Nov 20, 2002
Posts
7
Likes
0
I've seen the two resistors which set the output voltage on adjustable regs replaced by a diode and current source. What's the skinny on that?

Can anyone post a schematic and/or suggest part numbers?
 
     Share This Post       
post-222624
Post #8 of 10

ppl

Building amps and assuring water resistance.
Joined
Jun 21, 2001
Messages
1,772
Reaction score
13
Joined
Jun 21, 2001
Posts
1,772
Likes
13
See the att. The Zener is whatever output voltage you need plus 1.2 Volts. select the Jfets for an Current of about 10 Ma. MPF-103 I think is in that range. add a 1-2 ohm or less resistor between U1 out and Q1/C1 do the same for U2's out and Q2/C2. make sure the regulators inputs are bypassed as recomended in Data sheet or located real close the the Filtor caps.
 
     Share This Post       
post-233350
Post #9 of 10

puppyslugg

Headphoneus Supremus
Joined
Jan 31, 2002
Messages
1,642
Reaction score
10
Joined
Jan 31, 2002
Posts
1,642
Likes
10
For those having trouble locating the MPF-103 fet, try 2N5457. Mouser has the 2N5457's in stock.


CORRECTION:

It appears the cross referenced 2N5457 will not provide 10ma of sourcing. An option would be to use selected 2N5486's.
 
     Share This Post       
post-233663
Post #10 of 10

thegopher

New Head-Fier
Joined
Nov 20, 2002
Messages
7
Reaction score
0
Joined
Nov 20, 2002
Posts
7
Likes
0
ppl- is there any advantage in cascoding the current Source FET like you advocate for biasing opamps into class A?

If so, could you possibly post a schematic.

Many thanks

Gopher
 
     Share This Post       

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

Top