Speaking of Transformers...Ratio Question

[schmidlin]

This Q has been haunting me and I need to be straightened out.

The consensus: For OT's, the secondary drives the primary.  For example: I have an 8K primary into 8 ohms, if I change the load to 4 ohms, the primary is *forced* to 4K, causing inefficiency.  So they say.

So if I have a 150/10K input transformer, I could *force* it to be a 600 ohm primary by providing a 40K secondary load.  Or vice versa?

Is this how it works?  Is ratio absolute, like a gearbox?  Seems IT's are somehow different in this way.  Again, straighten me out.

Thanks in advance.

 

[rodabod]

So if I have a 150/10K input transformer, I could *force* it to be a 600 ohm primary by providing a 40K secondary load.  Or vice versa?

Is this how it works?  Is ratio absolute, like a gearbox? 

On a simple level, yes.

 

[abbey road d enfer]

Technically speaking, there is nothing like a 150:10k xfmr!
What it means is the transformer and its loads must obey the bridging rule (load Z>10xsource Z).
The primary inductance must be such that its intrinsic impedance is >1500 ohms, and the secondary load must be >100k.
As you know the intrinsic impedance depends on the frequency; the lower the freq, the lower the impedance. So all this must be seen in respect to the lowest transmitted frequency .
For 20Hz, the inductance must be >12H (1500/20.2pi).
If the inductance was half of it, the typical impedance would be 75 ohms, OR the lower cut-off frequency may be spec'd at 40 Hz.
The usage for most audio xfmr's is using 20Hz as basis of calculation, but there are exceptions: 70/100V loudspeaker xfmr's are very often rated for 40Hz cut-off.
Back to the OP, loading this said 150:10k xfmr with 40k at the secondary, would reflect 600 ohms at the primary for most of the bandwidth, except for LF where the primary inductance would reduce it slightly.
This is not taking into account the DCR of windings.

 

[PRR]

Transformer ratio is a lever.

But even for the same leverage, you use different levers for different work. I have a 7:1 lever for prying chips out of sockets, and a 7:1 lever for prying boulders out of the ground. Using the chip-lever on a boulder is futile, and the boulder-lever is far too clumsy and fat to fit under a chip.

> I could *force* it to be a 600 ohm primary

Well, almost. When the maker claimed "150 ohms", what he really meant was "at least 150 ohms from the lowest bass you paid for". Inductive impedance rises with frequency. A cheap 150 ohm at 50Hz iron won't get to 600 ohms until 200Hz, no matter what you do on the other side. Just like I can't exert 2-gram pressure with my 20,000 gram boulder-lever--- I'm fighting the lever more than the intended load.

Well, I'm abusing a simile a bit too much.

Get a signal generator. Put 2.2K and 820 in series across it. The impedance across the 820 is 600 ohms near-enuff. Connect this to your transformer, no load. Meter the voltage across the 820. Aim for 0.1V. Up around 1KHz, it won't change much with or without the transformer. At 100Hz the voltage will drop. When it drops to 70% (0.07V relative to 0.1V) the transformer no-load impedance is about 600 ohms, and your response in a true 600 ohm system will bass-cut lower frequencies.