Purpose of capacitors before transformer in the schematic

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

TijuanaKez

Well-known member
Joined
Jul 19, 2011
Messages
51
If I breadboard this design up as per the schematic from the Cinemag site, the signal sounds muffled which I assume is the result of the capacitors before the transformer acting as a LPF in conduction with the impedance at the input.

AN-108.pdf

So question is, why are they there?
 
No, that would make a high pass filter. Honestly I'm not sure what those caps are for. Keeping DC out I guess.

If you got the windings reversed it would probably sound muffled. Presumably that transformer is like 10:1 so if you reversed it, it would probably sound muffled.

Also the pad only works with low impedance sources. So not guitar. If you tried to play guitar direct into it with the pad in, aside from unnecessarily attenuating an already pretty low signal, it would also probably sound muffled.
 
If I breadboard this design up as per the schematic from the Cinemag site, the signal sounds muffled which I assume is the result of the capacitors before the transformer acting as a LPF in conduction with the impedance at the input.

AN-108.pdf

So question is, why are they there?
You should ask Cinemag, because it doesn't make much sense. It's apparently meant to protect against possible DC. Then why aren't they on the version that has no attenuator?
 
Even with the usual 2-3kohms of a modern mic input loading the transformer secondary it would still result in a fairly low load in 'instrument' terms . Wont be an issue if you have active guitar circuitry , an efx pedal or buffer in the chain but with a direct connection to a pickup you would expect a dull wheezy sound due to low load . Maybe you could try connecting it to a balanced line input(20kohms) with variable gain so the pickup sees a better load impedence.
 
Even with the usual 2-3kohms of a modern mic input loading the transformer secondary it would still result in a fairly low load in 'instrument' terms . Wont be an issue if you have active guitar circuitry , an efx pedal or buffer in the chain but with a direct connection to a pickup you would expect a dull wheezy sound due to low load . Maybe you could try connecting it to a balanced line input(20kohms) with variable gain so the pickup sees a better load impedence.
The xfmr is a 10:1, so the theoretical reflected impedance to the primary would be 200-300kohms, which, although being on the low side, could be acceptable.
The issue is that, in order to get a decent LF performance (I'm not talking about 20Hz, for guitar one needs about 100 Hz) the inductance of the hi-Z winding should be about 1000H, which I very much doubt in practice. Increasing the secondary load won't change much.
Actually that's a common issue with almost all existing passive DI's.
 
Thanks Abbey ,

So would I be right in saying with a given core material we'd do better with a higher ratio(more primary turns) like 30 or 50:1 in the case of direct pickup input ? but then in the case of lower impedence drive , like say a keyboard output we'd end up loosing out on HF performance due to the capacitance of all the extra turns on the primary side , in other words to provide reasonable performance under differing scenarios a compromise has to be struck .
 
Thanks Abbey ,

So would I be right in saying with a given core material we'd do better with a higher ratio(more primary turns) like 30 or 50:1 in the case of direct pickup input ? but then in the case of lower impedence drive , like say a keyboard output we'd end up loosing out on HF performance due to the capacitance of all the extra turns on the primary side , in other words to provide reasonable performance under differing scenarios a compromise has to be struck .
For a moment, forget about ratio. Just consider inductance. Unfortunately, mfgrs don't specify inductances (with good reasons). For a shortcut, let's use nominal impedance.
The high-ratio xfmrs you probably have in mind are moving coil cartridge xfmrs. They have a very high ratio because the low-Z winding has a very low nominal Z, typically a few ohms.
Their nominal Z on the high side is typically about 10k-ish. that means it will load the source (the guitar) just the same as yout 200:10k mic xfmr.
There are examples of 1:15, 1:18 mic xfmrs in vintage Teutonic tube preamps. These xfmrs were very expensive, requiring a lot of attention and know-how, that is, for all practical purposes, lost.
These xfmrs used chambered windings. You may get an approaching result in coupling several xfmrs with their hi-Z windings in series and their secondaries in parallels. Of course ther's a law of diminishing returns working against you.
 
If I breadboard this design up as per the schematic from the Cinemag site, the signal sounds muffled which I assume is the result of the capacitors before the transformer acting as a LPF in conduction with the impedance at the input.

AN-108.pdf

So question is, why are they there?
This capacitor is inserted for safety reasons to block DC an very low frequencies from the transformer. In many circumstances you don´t know how the ouptut of the internal guitar preamp is designed. The main problem is the complex resistance of the transformer. It isn´t only like a resisitor. There is additional capacity and also inductance. In some circumstances it is possible to get a vibrating structure.
The secondary problem is the reducing of signal-noise-ratio. So I advice you to use an preamp with a transformer output. In my view it is the best way to get a cool sound.
Best regards!
jokeramik
 

Latest posts

Back
Top