How to lower voltage rail a tiny bit

[Mikett]

I have a couple of transformers from an old amp I am thinking of using for a pair of LM3886 chip amps. If I remember the rails on the filtered output measures around 40-41V. This is after full wave rectification with a pair of computer grade 10,000 uF caps.
I know that the rails are a little high for a LM3886.

Can anyone give me some advice on how I can salvage these for this application? I was thinking of putting some rectifiers in series. Would these drop the voltage enough to a safe level?

What about just using a half wave rather than full wave rectification?

Thanks in advance for the responses.

[Joshua_G]

You can use power resistors between the trafo and the rectifier, and/or use CRC filter with adequate value of R.

[quasi]

Resistors will not work because when the amp is idling very little current will pass throught the resistors and very little voltage drop would occur.

Half wave rectification will not work either because at idle the supply capacitors will still charge to the peak voltage of the rectified AC waveform (minus the rectifier drop).

You require fixed voltage drop devices like series diodes for a small drop or a series regulator for a larger one. You could also use zener diodes to achieve the drop. Just calculate the current draw through the zener and multiply this by the voltage drop to select the power rating required.

Cheers
Quasi

[Ted205]

i think you would be best off regulating it down.

Pre-Regulator to drop to a safe voltage for further regulation. will require heatsinking

[Kashmire]

You can put a resistor in three different places, which all yield benefits.

1. Between the transformer and the rectifier (on the AC side)
2. Between the rectifier and the filter capacitors (on the DC side)
3. In series with the transformer's primary (on the high voltage AC side)

All of these options are similar, and deliberately "slow" the conduction times of the diodes, lower peak currents into the filter capacitors, and improve power factor.

How much average current does the LM3886 consume? Your peak currents will be many times this amount. How many volts do you need to lose?

Also, if the LM3886 operates in class AB, your power supply will have low current / high voltage when idle and "sag" under high demands. This is not good.

You can also try an Ametherm, which provides inrush current protection and soft-start. I use Ametherms on all my Pass Zen amplifiers.

You can also try a MOSFET capacitance multiplier, which loses 4V. Add a Zener diode, and it becomes a voltage regulator, but expect to lose a minimum of 6V. Try putting one of the 10,000uF in front of the MOSFET and the other behind the MOSFET.

[AndrewT]

Originally Posted by Mikett

I have a couple of transformers from an old amp I am thinking of using for a pair of LM3886 chip amps. If I remember the rails on the filtered output measures around 40-41V.

you have not mentioned the maximum and minimum output voltage over the range of mains voltage that your supplier can send.
Have asked your supplier what their normal tolerance on mains supply voltage is?

[MikeHunt79]

If you only need to drop down a few volts, you can always use diodes in series...

[macboy]

Lots of people speaking up but only suggestion so far is something I would personally even consider using: a capacitance multiplier. This will drop a few volts, and dramatically reduce ripple at the same time. Unlike a regulator, the voltage drop is fairly constant. That means that if the rail sags a bit, the capacitance multiplier continues to work exactly the same.
In addition to the link suggested above, check out: Capacitance Multiplier Power Supply Filter

[InlineTwin]

It is way simpler issue to deal with than adding components. You didn't mention what type of transformers they were. If they are toroid then it is easy to add a few windings opposite direction on the secondaries to lower it a few volts.

[Electrone]

If they are toroids, then I like InlineTwin's idea. Otherwise, I used to place small low voltage transformers in series with the large ones to boost the voltage. But reversing the winding polarity would then buck the voltage. The ampere rating of the small ones has to be the same or greater than the big one.