[LAU] OT: isolating transformers, redux

[Ken Restivo]

On Tue, Dec 28, 2010 at 12:10:39PM +0000, linuxdsp wrote:
> Ken Restivo wrote:
>> On Tue, Dec 28, 2010 at 11:12:48AM +0000, linuxdsp wrote:
>>> Ken Restivo wrote:
>>>> On Wed, Dec 22, 2010 at 07:49:26AM +0000, Folderol wrote:
>>>>> On Tue, 21 Dec 2010 22:10:07 -0800
>>>>> Niels Mayer <nielsmayer at gmail.com> wrote:
>>>>>
>>>>>> Jensen has some really good application papers talking about audio
>>>>>> transformers. By induction, one might get an idea of why they cost so
>>>>>> much -- same reason why other "pro" stuff with a reputable name also
>>>>>> costs a lot.
>>>>> <snip>
>>>>>
>>>>> I think some people are totally missing the point. The O/P has already said he
>>>>> doesn't want to spend a lot of money as the kit he's connecting to is only of
>>>>> moderate quality.
>>>>>
>>>>> I don't know if the links I suggested to Ken are any good (whether he's able to
>>>>> buy from those people) but I asked 'someone who knows'(tm) about these things
>>>>> and was told that the 'Oxford' transformers are more than adequate for the job.
>>>>>
>>>> Thanks!
>>>>
>>>> Yes, I obtained a pair of cheap telephone isolating transformers from Radio Shack. And it works! Yay! US$8 total.
>>>>
>>>> It is indeed specced for 300-5000Hz, but it seems pretty flat when run with a short (1 foot or so) wire. But, I'm noticing a lot of high-end rolloff with a 20-foot wire. Are there any components I could add to combat this?
>>> This is why you need transformers that are properly designed for the 
>>> use   you intend to put them to.  It might seem like a cheaper option 
>>> to buy  anything you can find that looks like it might work, but this 
>>> is often a  false economy.  It might be better to consider if the 
>>> problem you want  to solve can be achieved without resorting to 
>>> transformers of any kind.
>>> I suspect that you will find the response (even into a foot or so of  
>>> wire) is very far from flat.
>>>
>>
>> Well, fine, but, it's a moot point since I don't have any money. If "properly designed" == "expensive" (and it usually does), then it's a non-starter for me.
>>
>> I am trying to make the best I can out of the cheapest components available. So far, so good. This will do just fine. I'll experiment with some damping resistors and see how that works.  If not, good enough for now.
>>
>> Thanks everyone for the advice.
>>
>> -ken
>>
>
> You may be able to solve the original problem without resorting to  
> transformers at all, which might work out better, especially if funds  
> are limited.  IIRC you wanted to isolate the ground? Are you trying to  
> fix an earth / hum loop or some other issue?

Ah, thanks. The issue is ground hum, yes: hellacious, +10dB ground hum, louder than the program material. It has to do with an automotive application. +12VDC to the laptop (or +5VDC to iPod), +12VDC to the amp, and probably improper ground isolation within the laptop or iPod. 

I have had this problem before with laptops, but not as severe, and solved it by simply lifting the ground pin on the AC adapter. There is no AC adapter here so that's not an option.

But the transformer works well enough, so I'm happy.

-ken