trouble with LM1875-29V rails??

[rg24]

So I constructed the LM1875, boards from chipamp.com though purchased the parts on my own. I hooked the trans to the power supply and plugged it all in and am measuring 28.9V to 29V on the output rails as apposed to 25, this means that my 18V trans is 20V? I am a little confused, any help would be great. Please see photos for clarity. I have yet to connect the amp boards to the PS board because of this.

BTW I am running the 160V 18X2 trans and hope to power 8 ohm speakers.

Thanks

Ryan

[IMG][/IMG]

[AndrewT]

your 18Vac transformer only gives 18Vac when fed with the nominal voltage stated on the label and when loaded with a resistor that draws the maximum current stated on the label.

If the mains voltage changes then the output voltage will change pro-rata.
If the load current changes then the output voltage changes.

If you connect your transformer to your mains with no load on the output and measure the open circuit voltage and the mains voltage you should find that
Output voltage = Mains voltage/ Nominal voltage * 18 * transformer regulation.
A 160VA toroid will have regulation between 5% and 10%. EI is usually slightly higher.

20Vac open circuit with mains at nominal is actually a bit high, but I suspect your mains was running high at the time.

[rg24]

Thanks for your reply.

The mains are running at 120ish V here in my apartment in chicago. Though the trans is reporting an output of about 20.5V when hooked up to the ps (amp board are not hooked up) this give a voltage of 28 as apposed to the ideal 25V for 8 ohms. Is this just because its under no real load? I just don't want to fry the chips once I hook it all up.

Thx

[rg24]

The outlet voltage is 120V and the trans output is 20.5V when going into the ps (the amp boards are not hooked up). Is this because the amp boards are not hooked up? I just don't want to fry the chips as this voltage ends up becoming 28.8ish after the ps, which is close to being high for these little guys.

[AndrewT]

use a mains bulb tester to power up every new project and every modified project.

[djoffe]

Some assumptions...you're reading 20.5 Volts RMS at the output of the transformer. The caps charge up to the peak, which is 1.414 times the RMS. That gives us about 20.5*1.414=28.987. The lightly loaded diodes only drop a little bit, so your reported 28 plus a bit is quite plausible.

Max volts for the chip is +/-30, recommended operating point is +/-25. Probably, the power supply output will drop a bit once it's hooked up to the amp. The amp draws about 70 mA of supply current. That's probably enough to take the edge of the voltages, at least give you full diode drops, which should drop the output by 0.5 to 1 V.

AndrewT's series lightbulb is always a nice safety valve for new projects. Just hook a 25 watt 120 Volt incadescent lamp in series with the 120 volt side of the transformer. Often people will pull the fuse put the lamp across the now empty fuse holder.

If you still wanted to scrub off a little bit of voltage, you could drop a diode or 2 in series with the outputs of each rail (watch polarity).

Good luck...

[rg24]

Thats what I thought, thanks djoffe. I am actually currently working with danielwritesback on a revised PS that uses a one piece, square bridge rectifier to not only fix this issue, but also provides a cleaner signal. I will post my finished design so that all can see. He has been a HUGE help with this so far.

Cheers,

RG24

[rg24]

So with my revised PS the rails now measure about 28.3V each. They should drop 1/2V to 1V once it sees a load. It is now complete and I will be moving onto the amp enclosure next.

Here is a pic of the PS to date.
[IMG][/IMG]

[danielwritesbac]

Originally Posted by rg24

So with my revised PS the rails now measure about 28.3V each.
. . .

Ah, congrats! The tuner style snubbed rectifier has removed about 1v of peaky noise. That amplifier should run a bit cooler and thus somewhat more efficiently on its cleaner power. Mark Houston has that sort of filtering on his Synergy Gainclone project online. Synergy - LM3875 Gainclone Amplifier (Chip Amp) and there was the link to it.

As for the amplifier board, well thank goodness there's room underneath. As for now, I think it needs an input load--one of the three critically required resistors that are supposed to be attached directly to lm1875.
Here's a thread with a lot of information and quite a few decent LM1875 designs: LM1875 PCB, Which To Use

And, here's a link to the National Semiconductor lm1875 data sheet, directly from the manufacturer's site: http://www.national.com/ds/LM/LM1875.pdf

[thuytbd]

My Lm1875 amp is running at 30V per rail, but there is no problem at all.