It works well even at only 20 mA load. However, it is low dropout and high current design or Jung super regulator for power amplifiers. Your use case looks like preamplifier.
ATM, due the components shortages, there are 2 parts that are not easily available: SUP 70101EL MOSFET (available at Arrow) and DN2540G3 (available nowhere or maybe at some less known electronic shops).
For 100 mA and 24 V, if required several volts voltage drop is not a problem, there are other suitable regulators like Jung-Didden available in the diyaudio store:
Super Regulator – diyAudio Store
For preamplifiers or DAC, this regulator will be even better.
ATM, due the components shortages, there are 2 parts that are not easily available: SUP 70101EL MOSFET (available at Arrow) and DN2540G3 (available nowhere or maybe at some less known electronic shops).
For 100 mA and 24 V, if required several volts voltage drop is not a problem, there are other suitable regulators like Jung-Didden available in the diyaudio store:
Super Regulator – diyAudio Store
For preamplifiers or DAC, this regulator will be even better.
TO-243 has pins differently arranged. No easy way to connect this. Apparently, Mouser will have DN2540N3 in stock from November 4th.
At the project posting date, all parts were available, but just for several days. Another problem is that some company picked SUP70101 MOSFET for their design, at the same moment as I did and they are, from January this year, snatching any available stock anywhere in the world. They cleared 5000 pcs, from Mouser before they even reached warehouse, 1600 from DigiKey last month, 300 from Arrow … Damn.
There is no good alternative for this MOSFET. Mouser will receive them at end of December (hopefully).
If it were only DN2540G3 missing, I would add SMD option to the PCB design.
At the project posting date, all parts were available, but just for several days. Another problem is that some company picked SUP70101 MOSFET for their design, at the same moment as I did and they are, from January this year, snatching any available stock anywhere in the world. They cleared 5000 pcs, from Mouser before they even reached warehouse, 1600 from DigiKey last month, 300 from Arrow … Damn.
There is no good alternative for this MOSFET. Mouser will receive them at end of December (hopefully).
If it were only DN2540G3 missing, I would add SMD option to the PCB design.
There is a replacement in the BOM for this soon to be obsolete part. Real show stopper is P MOSFET SUP70101, which is unavailable for almost a year for reasons mentioned in the previous posts. There is no really good replacement and design was thoroughly tested, for stability and perfect performance under any conditions, with parts in the attached BOM (excel file). I would not recommend any MOSFET substitution before actually adjusting circuit compensation for this part and testing stability.
Circuit performs fabulous in my amplifier (LuDEF) and it is a pity that other members can’t try this regulator.
I have small stash of all critical parts (both MOSFET & DN2540) but I believe it is not simple procedure to send this to UK anymore (probably customs declaration and invoice needed?).
Circuit performs fabulous in my amplifier (LuDEF) and it is a pity that other members can’t try this regulator.
I have small stash of all critical parts (both MOSFET & DN2540) but I believe it is not simple procedure to send this to UK anymore (probably customs declaration and invoice needed?).
Any way to update with available substitutes for SUP70101? Agreed it may not be optimal, but some way to get most of the benefits?
I plan to make parts revision with available MOSFETs. But, it takes first to have anything suitable available. Current situation is frustrating.
For MOSFET replacement, thorough regulator testing and compensation adjustment is required for optimum performance. So, it must be done with chosen parts on the bench.
For MOSFET replacement, thorough regulator testing and compensation adjustment is required for optimum performance. So, it must be done with chosen parts on the bench.
I really like the idea of being able to drop a regulator into an existing design such as an older commercial amp.
Shame that some of the parts are now hard to come by. I see now even the TL431 has gone to 'long lead time' at Mouser.
Hopefully you'll be able to adapt the circuit to alternate parts, but otherwise I've bookmarked this one for future building when supply chains are back to some kind of 'normal'.
Shame that some of the parts are now hard to come by. I see now even the TL431 has gone to 'long lead time' at Mouser.
Hopefully you'll be able to adapt the circuit to alternate parts, but otherwise I've bookmarked this one for future building when supply chains are back to some kind of 'normal'.
Hi, can you talk about the permissible range of output load capacitance, over which the regulator remains stable and operates with good phase margin? Quite a number of power amp circuit boards include lots of local capacitance on their power supplies, which can be troublesome for some types of low dropout regulators with common source pass transistors. (the articles below, written by T.I. authors, for example)
Can you give a rough rule of thumb about how much load capacitance is okay, and how much load capacitance is too much? Thanks!!
_
Can you give a rough rule of thumb about how much load capacitance is okay, and how much load capacitance is too much? Thanks!!
_
Hi Mark,
regulators were severely tested with dynamic load using 470 uF/50V Panasonic FC as minimum and 68.000 uF (capacitors mix) as maximum. Stability was exemplary in those conditions. I don't expect any problems with 100 or 200 mF at the output. 220 uF as output capacitor was OK but on verge while applying 10 A transient load.
Stability is only affected by excessive voltage drop across the regulator due to Ciss/Vds variation and its effect on output phase. Actual compensation is targeted for stable operation with voltage drop from 0.3 V to. 10 V max.
regulators were severely tested with dynamic load using 470 uF/50V Panasonic FC as minimum and 68.000 uF (capacitors mix) as maximum. Stability was exemplary in those conditions. I don't expect any problems with 100 or 200 mF at the output. 220 uF as output capacitor was OK but on verge while applying 10 A transient load.
Stability is only affected by excessive voltage drop across the regulator due to Ciss/Vds variation and its effect on output phase. Actual compensation is targeted for stable operation with voltage drop from 0.3 V to. 10 V max.
Rick,
it was mentioned that hardest design goal to achieve was to make regulators tolerate HUGE capacitor banks at the output.
Thanks. Noticed already.
SQP120N10 and DN2540G3 are available as well. Only VOM1271 is missing, but there is a pin compatible replacement. Only question is whether there is still anyone that wants to build R21. If so, I can refresh parts list.
Midnight approaching here, so we can solve this tomorrow.
SQP120N10 and DN2540G3 are available as well. Only VOM1271 is missing, but there is a pin compatible replacement. Only question is whether there is still anyone that wants to build R21. If so, I can refresh parts list.
Midnight approaching here, so we can solve this tomorrow.
Very interesting solution.
For a new design with LT4320, would you still recommend a transformer with 2 V higher secondary voltage?
I'm not sure if the electricity here is stable enough for this solution. A local internet supplier blames it on low voltage from the supply line when we lose internet.
After rectifier and smoothing capacitor, average transformer used for an A class amplifier will provide DC output of about 1.25 x AC voltage. So, additional 2V will make + 2.5V. That’s about right for intended voltage drop on regulator. With LT4320, there is no voltage drop on rectifiers, so we get additional 1.5V. With stable mains, transformer with +1V is OK. In example, I have 500VA 2x19.5V transformer and regulators output is at 24.5V, with enough headroom for supply variation. I have nominal 220V varying from 212-227 V.
Using external heatsink for regulators, voltage drop can be, say 5V, what can cover large mains voltage variation (cca. 33V, but even higher AC output needed, +3V or so.).
Using external heatsink for regulators, voltage drop can be, say 5V, what can cover large mains voltage variation (cca. 33V, but even higher AC output needed, +3V or so.).