Linear PS for D class

Greetings and Salutations!

A noob question please:
I've got a huge toroidal transformer 2000VA 36V/55A just laying around.
Looking to build a basic linear PS to power two class D amps for a pair of bookshelf speakers and a sub.
Have some basic soldiering skills.
Can really use any advise on kits, parts.

Thank you
Dennis
 
Hi Dennis. Why do you want to regulate the input voltage of the class D amplifier? Class D amplifiers switch on frequencies far higher than the ripple voltage, hence power supply noise will likely not influence the amplifier modules.

36Volts rectified is about 55-58V under no load.

You are likely to experience more problems with your mains breakers tripping due to the input surge of a 2KVA transformer.

Anyway, Let us know what class D modules you plan on using what their maximum input voltage specifications are, and how much current they draw at maximum output.

Cheers,
V4LVE
 
  • Like
Reactions: 1 user
Disabled Account
Joined 2002
Linear power supplies are good in class D. I do so regularly as the results are often better than with the average cheap SMPS. However that is a very big transformer.

Some points of thought:

- you will need inrush protection! Fuse or breaker will melt/trip otherwise.

- 55A is serious current, don't make mistakes and have a fellow DIYer check before switching stuff on

- I take you will regulate the voltage? Then find an amplifier with right voltage rating. About 45...46V should be possible (regulated) staying within normal margins.

- when regulating make sure to properly wire and fuse the PSU. A DC side fuse is mandatory with such possible power/current. Even if the load is 2A the PSU can deliver much more when a short circuit occurs.

- When feeding 2 amplifiers calculate with 10A load current just in case so you will need a regulator regulating to 46V 10A. Of course you can calculate and build a 20A PSU if you like.

*Point of doubt: you mention "48V current" so there is now a red LED blinking....
 
Last edited:
Yes, a "red LED" is correct, its 55A current at 36V voltage output toroidal transformer.
I have no clue about building a PS, however "top level understood" all that you've written in your post. I've built few small kits before - a valve pre-amp, DSD DAC, D-class amp...

Would you suggest any readily available kits/schematics for a fitting PS with add-ons that you mentioned (in-rush protection + voltage regulator w fuses)?
 
Disabled Account
Joined 2002
Normally inrush protection is with the 2 kVA transformers. If it is in a casing please check. Normally it is a small circuit with a relay.

A 45V 10A Linear PSU is not that common today. Old linear telecom PSUs come to mind otherwise I think you'll need to build a discrete PSU.

Like you I have not built such a PSU but a quick check reveals this simple one. You would need just the upper half. I carefully avoid higher voltage PSUs to be able to use regulator ICs but LM723 can do this job.

https://www.homemade-circuits.com/0-to-50v-0-to10amp-variable-dual-power/
 
Last edited:
  • Like
Reactions: 1 user
99.99% of power amplifiers do NOT use regulated supplies .... think why.

A 2kW regulated power supply?
Does any of you two have any idea of what a dissipation monster that is?

Just do a little Math and calculate pass transistor dissipation and needed heatsinking.

As a side note, experienced Forum members should know better than suggest projects from no clue sites such as "Swagatam" "Home made circuits"
Same class as those suggesting "10 kW amps with 40 transistors per side". "1500W Powerful Bass Amplifier" powered by 3 transistors in parallel fed from a 12V alarm battery and similar wonders.
 

TNT

Member
Joined 2003
Paid Member
According to the seller both amps can work of a 48V current.
This is worrying....

A 36V AC trafo will be 50V after rectification.

48V is a voltage, not a current. When you write this kind of stuff, I feel I would like to advice you to not dabble with mains and power systems - at all - before you acquire som more insight. This is for you own safety. These are lethal things.....

The dimension seem grosly over-dimensioned for the amps you have in mind. Too big engine in a too small car....

//
 
  • Like
Reactions: 1 users
Disabled Account
Joined 2002
99.99% of power amplifiers do NOT use regulated supplies .... think why.

A 2kW regulated power supply?
Does any of you two have any idea of what a dissipation monster that is?

Just do a little Math and calculate pass transistor dissipation and needed heatsinking.

As a side note, experienced Forum members should know better than suggest projects from no clue sites such as "Swagatam" "Home made circuits"
Same class as those suggesting "10 kW amps with 40 transistors per side". "1500W Powerful Bass Amplifier" powered by 3 transistors in parallel fed from a 12V alarm battery and similar wonders.
These are class D amplifiers so they are power efficient. Regulation is not a problem with normal power levels in a home so a few Watt output power at most. The 2 kW is a maximum not the average power. One could use a 2 kW transformer to power a LED (with 22k series resistor and a 1N4148 in series) :)

It can be done and it will work even with relatively modest heat sinking. Of course it is way too heavy, it can be done with smaller transformers and I would indeed use a smaller one to make it a practical match. Contrary to classic amplifiers it won't pose the usual problems and have benefits. I have built several and they work just fine but it is clearly off the beaten path (which is cool).
 
Last edited:
Fully agree that the Class D amp itself will be very efficient and cold.

I feel mildly horrified by the inefficiency and HUGE thermal dissipation associated with a 2kW class regulated power supply.

Hence my suggestion to the OP to find an amp board which is happy with available, unregulated 50V+ supply.

Paying a little more for a suitable power amp will be much cheaper and more practical than down regulating to fit.

Take care.
 
Disabled Account
Joined 2002
There will be no huge inefficiency (look at class A!) and also no huge thermal dissipation. With MA12070 based amplifiers one starts to distrust the instruments. And what weight does mentioned ineffiency put on a scale when it are a few Watts or even a few tens of Watts lost to heat? The most popular section here has amplifiers that have the feature to burn much power and manage to play music too :)

With 7V differential (over the regulator) and 1A continuous current it will be 7W. Truth is that normal to even annoying loud levels in a home will result in currents below 1A and they will fluctuate. Otherwise I would not have chosen this. Large value filter caps preferably in CLC, low supply voltages, Schottky rectifiers, LDOs and modest output power make things possible. Nice stuff to experiment with.

As spoken of more often: picking/designing loudspeakers and amplifiers for the exact situation and the required power at ones own home has many benefits. It will give great satisfaction and makes one chuckle at square meter fan or watercooled amplifiers that are an expensive hobby with todays metal and energy prices.

The worries about 2 kW are not exactly justified. If they would be true one should also worry about the homes 16A 230V power when switching on a 3W LED lamp.
 
Last edited:
Disabled Account
Joined 2002
Yes. It can deliver 36V at 55A but it does not need to so it can be for instance 36V at 5A. Large yes, heavy yes, oversized yes, available for 0 Euro yes, usable yes. It will however be able to put out a more than usual high current when wiring makes a short circuit etc.

Here transformers above 500VA need inrush protection anyway (by regulations) but a 2 kVA one will definitely not work out OK without it.
 
Last edited:
  • Like
  • Thank You
Reactions: 1 users
With 7V differential (over the regulator) and 1A continuous current it will be 7W.
You joking???????

Your nonsense example is based on 1A Current (pulled out of the blue :eek: ) to calculate apparently low 7 W dissipation.

OP is talking about a FIFTY FIVE AMPERES transformer, so by your own example that would lead to SEVEN HUNDRED WATTS DISSIPATION!!!!

For fun try to design adequate heatsinking for that, how many power devices would be needed, etc.