MY DESIGN L20D IRS2092+IRFI4020H 200W8R

[lithoc]

My dad have a 48-0-48 and a 60-0-60 VAC.

It's not toroidal type but it's very heavy, >5kg per piece.

Which is more suitable for this amp?


@Bernie7 : 70VDC is not suitable because of the IRFI4020H can only produce 9.1amp. You need something higher or drop the voltage.

Refer to : http://www.irf.com/product-info/audio/audiofets.html

[trevmar]

Bernie7, IMO the T2 is really good enough for a quality audio system. It is clean enough so that it will be hard to find any speakers capable of distinguishing between the T2 and the L20D. However, the L20D measures cleaner, especially at high frequencies and high powers. I will use the T2 as a knock-about portable system run from my (Apple) 24V 2.65A power brick, and the L20D in the aluminum casing holding my main stereo amplifier. I want to get the SURE working for a suitcase-portable system I have designed, so I will get a real TDA7492 chip from Digikey or Mouser tomorrow, and replace whatever chip SURE supplied in the board

You know, the L20D performance is so good that it is really going to be hard to beat. Ljm_ljm has produced a board which is really stable, a very nice package indeed. And inexpensive as well It doesn't have all the protections of the T2 (speaker relays, etc) and that is just about the only way I think it could be improved. Other than increasing its power to kilowatts, of course

As for power supplies, you don't need as high a supply voltage to pump power into 4 ohm as you do into 8 ohms, but you need twice the current. The 4020 output chip on the L20D is optimized for high voltage, not high current.

As for pushing the specs a little, one tends to be cautious in view of the lack of speaker protection relay on the L20D. I am adding a board to do that for me But if you take 150 watts as the maximum real-world 4 ohm output (which is the number I would start with), you only need about 25 Vrms into the 4 ohm speakers to achieve that. You should be able to get that from a 40-0-40 power rail. Because it is a switching amplifier you can probably fudge it from a 68-0-68, but there will be higher peak VI values floating around, and the amplifier will not be operating at maximum efficiency. Look on pages 9,10,11 of the iraudamp7s data sheet and you will find curves using 50-0-50 into 4 ohms, for example (but they use a different FET in the lower voltage configurations than the 4020).

Hope that helps.
.

[Bernie7]

Hmm...table 1 of iraudamp7s data sheet says amp7s + 4020 does not support 4ohm loads. Does this rule out such speakers with L20D?

[trevmar]

I explained the issues with low impedance loads in my post above. IR took the attitude that if you didn't understand why there were issues, they were not going to bother with any explanation. You can see the curve I supplied for distortion vs power into 4 ohms, it goes out to 150 watts, but it was not taken with a 70-0-70 supply .

[Bernie7]

Ok gotcha about the specs.

What would happen if I used a 68-0-68 ps on a L20D? 200-300W demand would rarely ever happen and only on transient peaks, but if it does happen would the IR chip blow or just run out of breadth?

Reason I'm pursuing this is I already have a good linear Hypex 68-0-68VDC ps and using something else for my 4ohm speakers would be a deal breaker.

[trevmar]

If it were me, I would use it. The advantage of switching amplifiers is their ability to give good efficiency over a wide range of operating conditions Your 68-0-68 would probably drop a little under full load, and you very rarely would peak to full load in any case. Just be very careful that your 4 ohm speakers don't drop below that impedance, due to poor crossover design, or whatever. If you are driving a bi-amped speaker (active crossover) then the 4 ohm speaker will measure a much higher impedance throughout most of its range, in any case

But please don't sue me if the sub-$50 L20D boards can't handle the strain Expectations have to be realistic And don't forget to buy a speaker-protector relay board
.

[Bernie7]

Ok thanks for the affirmation. $50 amp modules are virtually disposable so I won't sue

But I assume the output won't clip at transient peaks as that would be really bad for my tweeters.

[trevmar]

That is a reasonable assumption. Remember that the output inductor averages the output current between the switch and your speaker, so if you aren't driving more than 150W into your 4 ohm speaker too often, you should be keeping the average switch current below 9 amps.

Mind you, I would like to hear what 150W sounds like into a tweeter. Let's see, 85dB efficiency at 1W, 125 dB at 100W, I guess 125dB is still reasonable
.

[Bernie7]

Btw my Hypex ps cards have a DC protection feature for speakers. Is there something else the speaker protection circuit should do, and can you give a link to a schematic? Thanks.

[trevmar]

There are basically the discrete circuitry protectors like this one:
1 set Speaker protection board DIY kit | eBay

and those using the NEC UPC 1237 chip, like this:
Speaker Protection Board Dual 16A Relay DIY Electron | eBay
Google for a datasheet on it

Both work well enough, be very careful to look at the power supply specs, some require a floating 12V AC power circuit to drive them.
.