winding impedance matching transformer

[bbksv]

I have 2 DVC (4 ohm) subs and a mono (500 x 1 at 2 ohms) amp. I was thinking about winding an impedance matching transformer to present a 2 ohm load to the amp...but I was wondering...would I need to make it with atleast a 500 VA rating?....if so...how would I do that?....is it all in the gauge of magnet wire I use?

[mwh-eng]

Sounds like you have two 4 ohms speakers and an amp rated at 500 watts driving a 2 ohm load. Put the 4 ohm speakers in parallel and you have 2 ohms of load. Besides, an amp rated for 500 watts at 2 ohms should produce 250 watts at 4 ohms. The impedance of a speaker is not constant with frequency.

I don't understand the need for a transformer.

[Alcaid]

He said it was DVC subs. Dual Voice Coil. That means 2x4ohm per sub. 4x4ohm in total. That can't be wired to 2ohm without a transformer. I can't help you on the transformer though..

[bbksv]

Yeah...I can only get 1 ohm or 4 ohm.....and at the frequencies I need them at ....they will be under 2 ohms...which will put the amp into a protection mode...

[sreten]

Simply put transformers and low frequencies don't mix well.

A 50/60Hz 500VA core will work at 50/60 Hz not surprisingly, but
would need derating for 25Hz, I think its by 1/4, if not its 1/2.

If its a class D amplifier I'd be surprised if the output into 2 ohm
is that much more than into 4 ohm to warrant the transformer.

sreten.

[mwh-eng]

Sorry I did not know what DVC stood for. Thanks for the correction!

You can use a 50/60 Hz magnetic core to go lower in frequency, but you will need more turns of copper to keep the core out of saturation at the lower frequencies. This is not a hard design, but the size of the transformer will be large and it will be heavy. You will need a laminated core. Magnetics Inc. is a manufacturer of the cores that will work in this application. They give away their design books and software that give you step by step instructions on how to design transformers. You can download all of this possibly.

Their website is http://www.mag-inc.com/

Another approach to this is to use a switching type of step down converter to change the impedance. This would require an inductor instead of a transformer, but the design is more difficult and you could have noise problems. I have not done a design such as this, but it can and has been done by others.

If you need help with designing a large and heavy transformer, I will look into it a bit. If you don't need to run your amp full volume you can get by with a design of lower than 500 VA. It's a duty cycle issue. If you run half volume you will need a 250 VA transformer. It can still provide the peak power of 500 VA, but the average would need to be 250 VA or lower. The lower your average power, the smaller the copper AWG requirement. You will still need the same transformer core area to support the applied volt-seconds in order to not saturate the core at peak power.

The transformer core will probably be fairly expensive for this project.

[sreten]

Originally posted by mwh-eng
You can use a 50/60 Hz magnetic core to go lower in frequency, but you will need more turns of copper to keep the core out of saturation at the lower frequencies.

I'm sorry but this doesn't make any sense to me, sreten.

[mwh-eng]

Not a very good explanation, I agree.

What I mean to say is the same type of core material and laminate thickness could be used. The amount of turns must satisfy the transformer design in terms of volt-seconds with regards for core area and Bmax, and not push the transformer into saturation at the worst case low frequency/high voltage magnitude. Obviously it won't work at DC and working at very low frequencies would create a very im-practical transformer. I have assumed this amp and transformer will only be used for low frequencies (something like 20 to 75 Hz), but this may be a false assumption. Due to core loss at frequencies higher than 60 Hz, a thinner laminate should be considered.

Using the same core material as used in a power transformer for 50/60 Hz, may not be the best material for an audio application. Due to coercive force, residual flux, etc., another material may be more suitable. Tube amp transformer core material and laminate thickness may be much better, depending on the frequency range.

I think a better solution would be to not use a transformer, but use speakers with the correct impedance. I don't know how much different speakers would cost, but a 500 VA transformer for operation down to 20 Hz will not be cheap to buy or make, IMO.

[ChocoHolic]

Hi bbksv!

500 W at 20Hz via transformer.....
You will get a big transformer.
And then assuming that the resistance of the speakers
voice coil will be more around 3 Ohms than 4 Ohms at these
frequencies + losses in the transformer ...
Then you migh consider that the amp might deliver 300W at
4 Ohms not only 250W.

All together you may gain from 300W in the speaker to 450 W in the
speaker, if you are lucky...
That's less than 1.8dB and you will not really notice an advantage
from all your efforts.
My proposal would be to connect the four voice coils in an 4 Ohms
configuration and be lucky.

Cheers
Markus