Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 25th April 2007, 05:48 AM   #1531
diyAudio Member
 
rjkdivin's Avatar
 
Join Date: Oct 2005
Linesource,
Your thumbnail link for the schematic is not functioning.
Robert
  Reply With Quote
Old 25th April 2007, 06:50 AM   #1532
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Hi Tomwaits,
the 317/337 regulator will work.
It is specified with the bypass transistor to increase current capacity and dissipation. For a front end supplying less than 40mA these add ons are not required.
I would be very tempted to keep a 33V Zener from input to output (without a series resistor) to help the output caps charge up and reduce the differential across the reg during start-up.

The 317/337 performance is pretty mediocre and gets worse at higher audio frequencies and terrible above audio frequencies.
The PSRR of the Krell will have to work hard to make up for this deficiency. Very good output caps can help reduce noise and impedance being fed to your front end and the reg keeps the low frequency spectrum in check.

A Pass style non feedback regulator MAY perform better than 317/337.
A discrete feedback reg should perform better after sufficient development.
__________________
regards Andrew T.
  Reply With Quote
Old 25th April 2007, 06:57 AM   #1533
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Default Re: Re: Regulator for Driver Board

Quote:
Originally posted by Crowbar

For a power amp, it's better to use a capacitance multiplier, as it allows you to maintain a minimum regulator dropout voltage without worrying about mains variation. I built Mr.Evil's design and it works good: http://mrevil.pwp.blueyonder.co.uk/a...sub/psu2.shtml
Hi,
has anyone tried a capacitance multiplier on JUST the drivers' supply. Even at 1.5Apk output the maximum dissipation will be just 5 to 6Wpk and tick over at about 300mW (could a FET be used as the active device irf540?).

Just had a look at mrevils circuit. I would never have realised it was a capacitance multiplier (well, activated version of). I see he uses a FET as the pass element.
NOTE, mrevil has returned the smoothing 0v separately to ground and NOT taken it to the reg output. This is the correct way to do it. One must strive to NOT contaminate the regulator 0v references with pulses coming into the smoothing.
__________________
regards Andrew T.
  Reply With Quote
Old 25th April 2007, 07:20 AM   #1534
PWatts is offline PWatts  South Africa
diyAudio Member
 
Join Date: Jun 2003
Location: Bloubergstrand
I've actually measured the LM317/337's to have better performance than many other IC regulators (not counting discrete ones) if properly implemented. Low-ESR caps on their outputs can actually lead to instability and it's more important to use a large output cap than a low-ESR one. At the load a nice one if parallelled with an electrolytic would help though.

The regulator I've been toying with is simply a high-power opamp with a stable reference (I used the 10V REF102) and gain. The OPA547 from TI is rated far higher than needed, and can supply a differential voltage of 60V. So you just need a high-voltage low-power opamp to perform the inverting function and run one 547 with only +Vcc and the other with -Vcc, configure the gain and there you have it. If you use the output of the positive regulator and run the inverting opamp and negative regulator at unity gain you even get the negative to track the positive but it's not necessary and I don't know of +-60V capable opamps.

I've tested one at 20W per 547 intended for a class-A preamp and it works excellent. Haven't used it in the actual circuit yet though but will soon. Of course since it's feedback-based it works best with the regulator located close to the load.

Since the highest-rated opamp I know allows +-45V rails it's only fit for the LTP which uses 39V, but if you can get a stable negative reference higher outputs are possible. For the drivers it may be necessary to go to the 548 or 549, the latter capable of 8A continuous and will handle the 3W or so used by the drivers without a sweat if you look at its SOA graph.

Unfortunately TI has recently stopped giving samples out of these excellent products and they're quite expensive.
  Reply With Quote
Old 25th April 2007, 07:22 AM   #1535
Account Disabled
 
Join Date: Mar 2007
Quote:
Originally posted by PWatts
Low-ESR caps on their outputs can actually lead to instability and it's more important to use a large output cap than a low-ESR one.
That is a simplification. I think it's worth considering this issue in detail. This paper is useful, and the measurement can be done with anyone that has a signal generator (use your soundcard only up to when its filter kicks in though, which will be well below 48 kHz for a 96 kHz sound card):
http://www.calex.com/pdf/3power_impedance.pdf
  Reply With Quote
Old 25th April 2007, 07:26 AM   #1536
PWatts is offline PWatts  South Africa
diyAudio Member
 
Join Date: Jun 2003
Location: Bloubergstrand
Of course it's a simplification (after all 99% of what's said on diyaudio is if you look at AES and IEEE papers on the subjects), but just to illustrate that adding exotic super-low ESR caps at the output of a 317/337 without damping isn't NECESSARILY a good idea and can have an adverse effect in certain instances.
  Reply With Quote
Old 25th April 2007, 07:29 AM   #1537
Account Disabled
 
Join Date: Mar 2007
I know, I was kind of looking for an excuse to include that paper, as a damping network as described to flatten the impedance vs frequency is a great way to improve a regulated supply.

Quote:
Originally posted by AndrewT
capacitance multiplier on JUST the drivers' supply.
If I were to use separate regulation for the rest of the amp (than the output stages), why not go all the way and use a good shunt regulator? Efficiency is really only an issue with the output stage anyway.

Quote:
mrevil has returned the smoothing 0v separately to ground and NOT taken it to the reg output.
Can you rephrase that? I'm not sure I completely understood. Transformer/rectifier ground is connected to the multiplier output ground--both have the ground symbol. What other 0 V do you see there?
  Reply With Quote
Old 25th April 2007, 07:33 AM   #1538
PWatts is offline PWatts  South Africa
diyAudio Member
 
Join Date: Jun 2003
Location: Bloubergstrand
And it was a good idea to put it here indeed Too many people believe that throwing money at something is bound to improve matters. This is not to discredit AndrewT's statement, as it holds true if implemented correctly.
  Reply With Quote
Old 25th April 2007, 07:52 AM   #1539
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Default Re: Regulator for Driver Board

Quote:
Originally posted by AndrewT
......Just had a look at mrevils circuit. ............NOTE, mrevil has returned the smoothing 0v separately to ground and NOT taken it to the reg output. This is the correct way to do it. One must strive to NOT contaminate the regulator 0v references with pulses coming into the smoothing.

Quote:
Originally posted by Crowbar
................Can you rephrase that? I'm not sure I completely understood. Transformer/rectifier ground is connected to the multiplier output ground--both have the ground symbol. What other 0 V do you see there?
I will try again.
They both show a ground symbol, indicating they are both connected somewhere else.
What is very important is that they are not connected from smoothing 0v direct to the nearest point on the schematic.
It is acceptable to take the smoothing 0v AND the regulator 0V using SEPARATE wires to a remote ground. This remote ground is often referred to as star ground, audio ground, main ground, etc. (but not chassis safety earth).
__________________
regards Andrew T.
  Reply With Quote
Old 25th April 2007, 08:08 AM   #1540
AndrewT is offline AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Quote:
Originally posted by PWatts
I've actually measured the LM317/337's to have better performance than many other IC regulators (not counting discrete ones) if properly implemented. Low-ESR caps on their outputs can actually lead to instability and it's more important to use a large output cap than a low-ESR one. At the load a nice one if parallelled with an electrolytic would help though.
The datasheet for 317 & 337 show a low esr on both the input AND the output to ensure stability.
The instability you refer to is more usually associated with LDO regulators. I was not aware that the stability region diagrams afflicted the conventional 317/337 types (they are NOT shown in the National datasheets).

I read very recently that some 317s are available as low drop out versions and that these can fall into the normal LDO stability problems and low ESR aggravates this.

Can you confirm which types of 317/337 you are specifically referring to?

Read National.com, lm317, page 8, external capacitors. It even warns that some ceramics have a capacitance decrease at 500kHz and that this failing is overcome by swamping the effect by using lower impedance caps (1uF solid tantalum - but not at 60Vdc) on the output. (this is not a failing of the reg, it is the ceramic failing National are giving the solution for). The fact that National even mention ceramics on the output implies they are approving the use of ceramics at this location.
__________________
regards Andrew T.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



New To Site? Need Help?

All times are GMT. The time now is 10:45 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2