DIY stereo amplifier with laptop charger

Hello all,

If we build a DIY stereo amplifier class AB, and if we use a regular Laptop Charger as power supply, what is the maximum power (watts) per channel we can get ???

Secondly, is ther any disadvantage in terms of audio quality by using laptop charger as power supply ??

Thanks in advance,
Naveen
 
Hello all,

If we build a DIY stereo amplifier class AB, and if we use a regular Laptop Charger as power supply, what is the maximum power (watts) per channel we can get ???

power == sqr(Vrms) / load

If you run it single ended ...
approximately (19v / 2) X .707 == 6.7 volts RMS
(6.7 X 6.7) / 8 == 5.6 watts.

if your run it in BTL mode ...
19v X .707 == 13 volts RMS
(13 X 13) / 8 == 22.5 watts.

Secondly, is ther any disadvantage in terms of audio quality by using laptop charger as power supply ??

If you mean "external SMPS power supplies" my answer is "Yes".

The thing is that SMPS (Switch Mode Power Supplies) work at relatively high frequencies and are tightly regulated. So you are getting high purity DC that does not ripple or hum under load.
 
Member
Joined 2010
Paid Member
Laptop power supplies are generic quality SMPS devices with an emphasis on reliability - but perhaps only for a few years in constant daily use. They can work fine with audio devices provided that there is adequate input filtering to reduce the likely high frequency switching artefacts remaining on the DC. There are already quite a few HT amplifiers and good quality amplifiers of class D, AB and even a few class A amplifiers that come with SMPSs and have been for quite some years. There aren't many reasons why you can't use them other than the limitations of their power rating and single polarity, fixed 19V supply.

However, they do generate a lot of switching noise and it's hard to attenuate once it enters the metal case of even a well constructed amplifier. Filtering the DC output first, can begin with something as simple as clipping a large ferrite suppression bead on the cable or better, by constructing a proper low-pass filter and also fitting that before it enters the amplifier's metal shield case. Filtering the HF hash of a SMPS
 
However, they do generate a lot of switching noise and it's hard to attenuate once it enters the metal case of even a well constructed amplifier.

That's why you use an external supply. The switching noise is thus, not in your equipment's cabinet.

Take an RF Sniffer and scan it around the newer SMPS bricks... they're pretty darned quiet these days.
 
Definitely i will drop the idea of using a laptop charger. I'm looking forward to get a DIY amp made with around 50w per channel. And i don't want to compromise on audio quality.

Thank you so much all of you for detailed explanations !!!!

A 24 volt supply running BTL amplifiers (lots of AB and D chip amps around for that) will give you a very reliable 35watts on 8 ohms or 50 on 4.

For example... THIS little guy will give you a very clean 35 watts per channel, all day long, at about .01% distortion with a -3db response from 8hz to at least 30khz. And it runs on a 24volt brick.

The audio world is changing, good audio is getting cheaper... and cheap audio is getting better.
 
Member
Joined 2010
Paid Member
I think the 50W amplifier the OP is planning on will likely be a traditional, discrete class AB DIY design that needs around +/- 30V supplies for 50W/8R. A dual rail SMPS is certainly doable with some adjustable modules that are available online but this opens a safety issue, covered generally in note 3 of the forum rules. What do you say, Naveen?
 
I built my 2 separate power amplifiers back in 1982, Mosfet type, each rated 112W in 4 ohms. They have worked flawlessly and still do, but recently one of the toroidal transformers started humming (mechanical, internal) and I was not able to do anything about it.

I decided to replace the transformer and rectifier with a set of 6 compact SMPS laptop supplies I had laying around. I left my capacitor bank in place (2 x 4 x 4700 uF) .

The power supplies I used are 19 V 3.3A but anything above 2 A would have been enough. I connected all the outputs in series which results in a dual rail supply of + and - 57 V.

It works like a dream. I think this is a good way of re-using obsolete supplies which are easy and cheap to find. As for safety, I personally have no doubts. These supplies are normally thermally protected, short circuit protected, current limited and deliver very stable output voltage. The efficiency is high so they do not produce a lot of heat anyway. My old transformer supply was unregulated, this must be a lot better.

I can not hear any sign of switching interference with the amplifier itself, nor any sound from the SMPS supplies.

I do think it is important to use supplies from a well known brand. For safety reasons I would not use cheap Ebay supplies for this, since they might not be protected as good as the one that came with your laptop.

Another important thing: your amplifier must have a protection circuit that disconnects the loudspeakers when the power voltages are not within range. When turning on the system the 6 supplies will react differently on the high capacitor load so it might take some seconds before all supplies are happy and both positive and negative sections have reached their specified voltage. In my case it takes around 1-2 seconds longer before I hear the well known sound of the loudspeaker relay clicking in. After that you are good to go and I do not experience any power cuts during high system load.

The amplifiers: ALBS 100 N
The loudspeakers: Quad ESL 2805
 
For a single one-voltage supply as for powering a laptop, I really like the idea of going BTL as opposed to single-ended output, as the major currents will be pulled directly from the power supply, and the artificially-generated ground will have much lower currents going through it, so the ground generation circuit won't need nearly as much current capacity as would a single-ended output.
 
For a single one-voltage supply as for powering a laptop, I really like the idea of going BTL as opposed to single-ended output, as the major currents will be pulled directly from the power supply, and the artificially-generated ground will have much lower currents going through it, so the ground generation circuit won't need nearly as much current capacity as would a single-ended output.
I am not sure I understand your comment, could you elaborate on it?