1800VA Torroid to get 40A???

[Dominick22]

So, I have learned that my Martin Logan speakers like alot of current, so I bought a (what I thought) was large 420 VA 22V transformer which delivers 9.5 Amps pre rail.

I just found out that the ML's like amps that run around 40 Amps per rail. Always, with the revisions!

So now I am thinking that to get this, I would need an 1800VA torroid @ 22V or a 1600VA @ 20V.

But, I also know that the primary Ampere input can affect the overall output...this is my inquiry...

1. How do I figure the primary Ampere input and its effect on Ampere output?

2. If I were to find an 1800VA @ 22V Torroid, would I really be capable of pushing 40A with the LM4780? What is the limit? 35V is the only restriction I have ever seen quoted on here.

3. Would the 1600VA @ 20V or 1800VA @ 22V be the better choice?

Thanks,
Dominick

[otherside]

In an ideal world it would be:

Vpri x Ipri = Vsec x Isec

However all transformers have a level of effiency usually stated on the datasheet. The closer the output voltage to the input usually the higher the effiency. on 230V->22V efficiency is around 90%, i'm not sure about 110->22V. Have a look on the datasheet.


So the Equation becomes:

Vpri x Ipri = (Vsec x Isec) / Efficiency. Well the exact formula is a bit different but this is the simplest way to do it.

So the primary current would be Ipri = (Vsec x Isec) / (Efficiency x Vsec).

If for example you have a 1800VA transformer, primary at 110v and eff around 91%:

Ipri = 1800/ (110 x 0.91) =~18A.

I'm not sure about the US standards but in europe usually sockets are connected to 16A fuses... Anyway you would need quite a beefy cable for it (at least 4mm) on the primary (normal power cord wouldn't be enough).... and double (not exactly) on the sec.

What Amp are you using ?

BTW the cables to your ML speaker should be quite beefy too!

[Dominick22]

I just finished an LM4780 design that you can peek at in this section. I guess the sunfire grand cinema pushes 40A per channel and current is the hot commodity for ML's so I thought I would make that rating my goal.

The sunfire is also a favorite among ML owners.

What does the Ipri, Isec stand for? I guess just the I is unknown to me!

Thanks for the help,
Dominick

[otherside]

I : current (ampere)
V : voltage

Ipri : current on the primary winding (transformer input)
Isec: current on the secondary winding (transformer output)

Vpri : voltage on primary (110 in US)
Vsec: voltage on secondary (22V in your case)

[AKN]

According to LM4870 datasheet chip can supply a maximum of 11.5A if not limited by SOA. To me that big transformer should be of waste.

[Dominick22]

Thanks,

That is what I thought somone might say. So 9.5 has the limits almost maxed as it is. I guess I will just have to be happy the way it is.

Thanks otherside and 4fun

[otherside]

A small correction to the stuff i wrote above, efficiency improves with size not output voltage. A huge tr i guess its possible to have close to 95% efficiency.


BWT you could parallel up (based on the design) 4 LM4780 to get 40amps out .

[AndrewT]

Hi Otherside,

If for example you have a 1800VA transformer, primary at 110v and eff around 91%:

if this is true then throw it in the bin.

A good small transformer of about 300VA should achieve a maximum efficiency of about 95%, by the time you reach 1800VA the maximum efficiency will be approaching 98%.

[AndrewT]

Hi Dom,
I think Martin Logans need amp performance that chips will never be able to deliver.

But back to your query.

The transformer charges the smoothing caps. The designer selects a transformer Vac and VA to suit the maximum average power the rails will need to supply for normal operating conditions.
For a normal ClassAB amplifier it is usual to specify a VA rating equal to 1.5 times the maximum power that you can push into your normal load.

The capacitors supply the transient current to the output stage. This transient current can be many times the average current that you charge the caps with.

Some well designed small amplifiers can achieve large peak current ability. I do not know the US market, but here in the UK the audio press and the better manufacturers specify the peak current ability of the amplifer.

An example would be a 50W into 8r amp that can do 25Apk.
This peak current is about 10 times the rms current into an 8r load.
Even the peak current into an 8r load is only 3.5Apk
But the specification confirms that the output stage and Power Supply can achieve short term transients that very demanding speakers can require.

I am confirming that it is possible to design an amplifier that can successfully drive the Martin Logans, but you do need to know what you are trying to achieve and how to get there.

[Eva]

There is not such a thing as 40A instantaneous currents being sourced by amplifiers or flowing through a loudspeakers while listening to music.

The only recommendation that I can make is to place a precision non-inductive current shunt in series with the loudspeaker and to monitor the current (voltage drop) waveform across it with a suitable oscilloscope (be careful with earthed equipment and ground loops).

Then try to reach just 10A transients with music, without causing the amplifier to clip, and without blowing the loudspeakers, your ears or both...