SMPS problem?

[choncba]

Hi, I recently finish a car amplifier based on two LM3886 chips working in parallel mode. I build the SMPS whit SG3525 PWM chip, driving four IRFZ48 mosfet, and the transformer have 8 turns primary and 24 turns the secundary (6mm total coil). The schematic is in this forum. When I put that on maximum power the +/- 35 Vcc falls too much (to +/- 26v ). Its obviusly reduce the output power. The 12v cable have a 4mm diameter.
Any body have that problem too???
How can I fix that?

Thanks you.

PD: Sorry by my English (I AM from Argentina)

[Perry Babin]

If you want the voltage to hold precisely at ±35v, you'll need to use regulation.

If you don't want to use regulation, you would need to provide much more information about the power supply.

Core material?

Core size?

Operating frequency?

DC input voltage no load?

DC input voltage with load?

Primary wire size (number of strands, diameter of each strand)?

Secondary wire size (number of strands, diameter of each strand)?

[luka]

Hi

My first guess is that your input voltage is not constant and 2nd you have unregulated supply, so you can't have constant output voltage...

[choncba]

Originally posted by Perry Babin
If you want the voltage to hold precisely at ±35v, you'll need to use regulation.

If you don't want to use regulation, you would need to provide much more information about the power supply.

Core material?

Core size?

Operating frequency?

DC input voltage no load?

DC input voltage with load?

Primary wire size (number of strands, diameter of each strand)?

Secondary wire size (number of strands, diameter of each strand)?

Thanks for answer, here are the data:

Core material: ferrite

Core size: 3 colums type, 50mm*50mm 10mm diameter

Operating frequency: 54Khz

DC input voltage no load: 12.8v

DC input voltage with load: 12.3v (full load) -> engine running

Primary wire size (number of strands, diameter of each strand):

Primary wire is 2.4mm with 8 turns, central tap.

I now that the outputs are unregulated, but i expect better results!

The otutputs capacitors (filters) are 2*(4700uF+2200uF) +/-35v



Thank you.

[Perry Babin]

When I asked for core material, I was referring to the composition of the ferrite material. For example, F and P materials are used for power transformers. Other manufacturers use numbers (77?, 75?) to specify the material.

I don't have enough experience with those cores to be of any help. If no one replies here, try posting this on the power supply design forum.

If this was in a vehicle with the engine running, you either have a defective alternator or the power wire was too small. With the engine running, you should have had 13-13.5v when loading the supply. Approximately 1.5-2 volts of loss per rail is due to the B+ voltage drop.

Are you driving a 4 ohm or an 8 ohm load?

[DigitalJunkie]

DC input voltage with load: 12.3v (full load) -> engine running

Was this at the amplifier end?
What is the voltage at the battery terminals with the engine running?

It could be voltage drop across the positive wire,if it's too small. (this could account for the sagging rails a bit also)

Otherwise if that's the voltage at the battery,Perry is right,Your alternator or something is shot.

[choncba]

Sorry, the voltage from battery is OK, was an error on measure. The value is 13,4v with engine running. I drive a GTT300 blaupunkt's bazzoka whit 12 inch woofer, at 4 ohms. The amplifier work's with two LM3886 IC, from national. The problem is that power supply gone down at full imput signal, so the output never have 100 W RMS. I need to modify something on the SMPS?

[Perry Babin]

You'll probably have to add regulation to your supply. To allow the supply to hold the target voltage (±35v), you'll have to add windings to the secondary.

If you have ±35v now with the amplifier connected but with no audio output, you should try increasing the secondary to 34 turns (from 24 to 34). That should get you close (it should hold to ±35 at full power).

You need to disconnect the amp while testing the regulated supply. Unregulated, it will produce more voltage than the amp chip can handle. If the regulation isn't working properly, you could blow the amp. After you get the regulation working properly, then you can reconnect the amp.

[choncba]

Thank's for the answer, I will try to use regulation and modify the secondary of the transformer, but, what kind of regulation you recomend? I'm gonna try using TIP3155 transistors, or similar, what you suggest?

[Perry Babin]

You stated that you used the schematic from this forum. I assumed that you meant the one in the sticky at the top of the page. If that's the case, it included regulation. If you used a different one, you need to post a link to it.

You won't use linear regulation (transistors passing voltage from the rails). The series pass transistors would dissipate too much power. The regulation will be through the switching control IC.