Transformer Mojo box frequency response

[mattenylander]

Hi
I’m fairly new to electronics but I’ve gotten interested in some passive transformer coloration that I could add to my vocals chain.

My goal is to be able to make a transformer box that I can patch in between say my preamp and to my Apogee Duet interface so I can print the effect so to say.

I’ve tested the transformer in post, by patching Duet out -> transformer -> Duet In.

I’ve also patch it Duet out -> Sub Zero RS100 preamp -> transformer -> Duet In

And also different if I go Duet Out -> RS100 preamp -> warm audio Bus Comp -> transformer-> Duet In

But when I measure the frequency response via the transformer the level and the frequency response vary. I mainly see a bass roll off so my question is,
what’s causing this bass roll off?

And are there any circuitry I can make to make make up for it? And to make it more suitable to be patched in between my gear?

Here is the response


And here’s the little transformer

 

[NOON]

There are others on here that are far more experienced with transformers than me, but here's my take. That's a reasonably appropriate curve for vocals but wouldn't want it over a whole mix. Could just be that transformer, but low end response can be affected by driving impedance. Because of inductance the impedance of the transformer is less at lower frequencies, forming a divider with the output impedance of whatever is driving it.

If your preamp has an output impedance of 50 ohms (fairly typical) then by the time the impedance of the transformer reaches 50 ohms you're halving the signal or 6db down.

Transformers made for this kind of purpose are usually bigger with a higher inductance at low frequencies so they're less affected by output impedance. The remedy is get a different transformer or eq some bottom end back in.

 

[Newmarket]

Do you have the transformer spec ?

 

[mattenylander]

Because of inductance the impedance of the transformer is less at lower frequencies, forming a divider with the output impedance of whatever is driving it.

Ok so what attributes in the transformers are mainly causing it’s inductance value? Can I alter it somehow?

 

[mattenylander]

Do you have the transformer spec ?

Unfortunately not… I picked it from an old cassette player I think.. I measured it with a multimeter and I remember correctly it had 8 Ohms on side and about 3,4 KOhm on the other side that I’ve connected it to.
Can I measure the inductance of It somehow?

 

[Newmarket]

Ok so what attributes in the transformers are mainly causing it’s inductance value? Can I alter it somehow?

Core Material and number of windings basically (I'm sure others here with far more knowledge could elaborate).
I don't see that it's practicable to modify that transformer.
As NOON said - it looks as if it might be fine for vocals - you can always add a bit of EQ to the signal to taste.
Do you think it's benefitting your vocal chain ? (Assuming you've tried it on actual vocals)
If you want significant 'colour' then hit it hard !

 

[mattenylander]

Ok thanks! Then I’ll go and accept that little bit of roll off.
Are there any ways I can make it match better with my other gears impedance? Since I saw a different kind of level drop patching in/out from my gear in various combinations,Like a potentiometer at input/output of the Mojo device?
Yeah I’ve tried it on vocals and think it adds a really nice distortion/grit in the bass region which I like.

Here’s what harmonic it adds to a 100Hz sine what it adds,
With out transformer,
With transformer,

 

[NOON]

That's an entirely inappropriate transformer for clean and hifi audio in that situation, but that doesn't mean it's wrong. If what you're after is distortion then something entirely inappropriate is a good starting point. Try it both ways around, it will step up and step down but respond to your other gear differently. Load it down with a pot so you can get overdrive it, then pad it down. Put caps in series with it and you'll get resonances at different frequencies. If you're not going for flat and clean then just try stuff and watch what happens on your analyzer.

 

[ruffrecords]

The main factors that affect the distortion produced by a transformer are:

1. Signal level - the bigger the signal the more the distortion
2. Frequency - transformers distort more at lower frequencies
3. Core material/size. The bigger the core cross section the more flux is can take and hence the more signal before it distorts.
4. Driving source impedance. The higher the driving source impedance the more distortion the transformer will produce

Strangely, load impedance does not really affect distortion but it does affect frequency response.

Most of the above factors are fixed by the transformer you have but for added colouration you could try adding a series resistor to the driving source. It is hard to say what value to use without knowing the transformer specs but you could try 1K, 5K and 10K and see what happens.

Cheers

Ian

 

[CJ]

check the bass response at a lower input level,

 

[mattenylander]

check the bass response at a lower input level,

I checked it, but I couldn’t measure any noticeable difference…

 

[mattenylander]

Driving source impedance. The higher the driving source impedance the more distortion the transformer will produce

Thanks! Ok I understand, I actually got a couples of Carnhill VTB 2281 yesterdays which I’ve hooked up and I notice the levels are “all over the place” depending on where in my chain I patch them in. Especially if I hook the VTB 2281 Transformer after the Transformer output on the Warm audio bus comp then I also noticed that the, transformer-> transformer configuration cause an even bigger low Frequency roll off, what’s causing this increased roll off? Increased inductance?
How would one design in general to match output // input impedance from one stage to a transformer?

 

[Silveradodays]

Hi,

You can extend the low freq response by adding a fairly big capacitor (4 uf, 6.8uf, 10uf ...) in series with the primary. It changes the slope of the high pass filter (2nd order filter vs 1st order) but extends the cut off frequency accordingly.
Depending on the impedances of the surroundings and the resistance of the xformer you might even get a "hint" of resonance at the cut-off which might be cool for color.

 

[Silveradodays]

See this real world example with the ultra cheap Triad Typ250, you could extend the highs in the same way by adding a small inductor in series too:

 

[mattenylander]

you could extend the highs in the same way by adding a small inductor in series too:

Thanks! Really nice! I didn’t have to many options of capacitor values at home but I had a 10 uF cap that I put In series on the primary as you suggested, and then I took like 80nF cap in parallel on the secondary which gave it a nice smiley curve

 

[Silveradodays]

awesome!

 

[ruffrecords]

How would one design in general to match output // input impedance from one stage to a transformer?

These days you don't really match transformer input and output impedance. As I have said many times before, transformers do what they say on the tin. They transform voltages, currents and impedances. They have no intrinsic impedance of their own. However, the engineering limitations of making real transformers means that the transformations will only take place across the entire audio spectrum for a limited range of impedances and it is these impedances that are quoted by the manufacturer. In that sense they are really recommended operating impedances.

The VTB228 you have is nominally a 600:600 transformer. What this means is that the driving source impedance should be no more than 600 ohms and the lower the better, Any higher and the inductance of the primary winding will end to cause a loss at low frequencies. It also means the load impedance should be 600 ohms or higher. Any lower than 600 ohms and again you will lose low frequency response. Higher than 600 ohms is fine.

For an excellent introduction to audio transformers I recommend you read this:

https://www.jensen-transformers.com/wp-content/uploads/2014/08/Audio-Transformers-Chapter.pdf

Cheers

Ian

 

[ach]

In that sense they are really recommended operating impedances.

I would even go so far as to say that they are simply 'The operating impedances at which our specs are valid'.
I consider most (vintage especially) transformer specs as simply usage indicators and a way to calculate approximately the missing factors (like inductance).
Oddly, while in general it is the least important spec, I often get the best quick indicator for appropriate usage from the winding DC resistance value – as they give a good 'magnitude approximation' of expected inductance without knowing anything about the core.