Keyboards, Interface, and DI

bdam123

Member
So I have a number of keyboards that I want to hook up into my PTHD setup. I previously had plans to route all of them into directly into my interface after they hit a Coleman MS6 Aux Switcher (I have 6 keyboards that will be routed into a pair of inputs).

Then it dawned on me, would the keyboards sound better if I ran them through a nice quality DI first?

While searching for an opinion, I came across this article.

In short, it talks about "untested" keyboards having the potential of damaging the line input op‑amp of the interface if plugged directly in.

My question is, how do you test if a keyboard is safe to plug in direct?

Also, what are you guys' opinion on recording keyboards? Direct or DI?
 
I will admit, I have no idea what you are talking about. How can an output from a keyboard damage your interface, it would also damage your mixer I would think? You do have to be careful of your output setting, and if the keyboards are really crap, guess they could kill your interface, but I wouldn't think any main brand keyboard would do that.

For hooking up, I would think either a multi-channel interface or a mixer for all of the keyboards and the two outputs to the interface. Since there is just one player, that would be the way I would do it.

Don't forget your MIDI routing!
 
Hugh, who wrote that article knows his stuff. I think the points are valid, if perhaps a little rare.

Most pro keyboards are perfectly safe to connect, and as he says himself, he's connected them that way - but don't forget that some on the road for ages type keyboards may well have all sorts of bodges applied internally, as needs appeared. Hums on outputs 'solved' by the owner opening up the keyboard and extra ground wire run from the power supply, or perhaps a keyboard where the headphone output has been the only way out that the device has, so somebody opened it up and soldered a direct line out to the board before the headphone amp, unaware that at that point, the audio had a DC component on it. Often guitar amps have DC blocking capacitors on the inputs, as do interfaces that have phantom power, so the problem keyboard may get away with it until plugged into something less happy.

All rare things, but I've seen them - and while I am happy to connect my gear to my interfaces, would I risk that battered old keyboard? Unlikely, but that old keyboard is also likely to be unbalanced, and a DI is a nice simple isolation device.

Apart from the higher input impedance which reduces loading, the main benefit for me is just convenience and a bit of extra isolation.
 
If there's DC on the output, you can measure it with a meter. Plug in a cable put one probe on the sleeve and the other on the tip. It really doesn't even matter which probe is which. If the absolute value of the reading is more than almost nothing, there might be a problem.

Actually, though, most well-designed equipment nowadays will have DC blocking caps at both ends anyway, so unless the voltage is either bigger than the V rating of the cap or bigger than the bias voltage on the opamp, it's mostly just about pops on plugging/unplugging.

Even in the extremely rare case that your interface has a split supply and is DC coupled to the outside world, a bit of DC offset on the input won't fry the opamp unless it exceeds its rates input differential max. It will eat headroom and cause it to clip assymetrically, but it would take some pretty big voltages to actually hurt your interface in most cases.

I still wouldn't want to see anything near a volt at that point, but... Remember to let the voltage reading on your meter settle before freak out.

I tend to think that if you've got significant DC on the output of a keyboard, the right thing to do is fix the damn keyboard. I stay away from passive DIs because they always cause a voltage step down, which compromises S/N. They can block DC, and help break ground loops. Use them in a pinch to get you through a gig or idiot proof a quick set up, but do not rely on them for (semi)permanent installations. Fix the actual problems instead. A 1:1 isolation transformer is a bit better solution, but transformers are not actually flat or linear. Some folks dig that coloration, but I prefer straight wire whenever possible.
 
If there's DC on the output, you can measure it with a meter. Plug in a cable put one probe on the sleeve and the other on the tip. It really doesn't even matter which probe is which. If the absolute value of the reading is more than almost nothing, there might be a problem.

Actually, though, most well-designed equipment nowadays will have DC blocking caps at both ends anyway, so unless the voltage is either bigger than the V rating of the cap or bigger than the bias voltage on the opamp, it's mostly just about pops on plugging/unplugging.

Even in the extremely rare case that your interface has a split supply and is DC coupled to the outside world, a bit of DC offset on the input won't fry the opamp unless it exceeds its rates input differential max. It will eat headroom and cause it to clip assymetrically, but it would take some pretty big voltages to actually hurt your interface in most cases.

I still wouldn't want to see anything near a volt at that point, but... Remember to let the voltage reading on your meter settle before freak out.

I tend to think that if you've got significant DC on the output of a keyboard, the right thing to do is fix the damn keyboard. I stay away from passive DIs because they always cause a voltage step down, which compromises S/N. They can block DC, and help break ground loops. Use them in a pinch to get you through a gig or idiot proof a quick set up, but do not rely on them for (semi)permanent installations. Fix the actual problems instead. A 1:1 isolation transformer is a bit better solution, but transformers are not actually flat or linear. Some folks dig that coloration, but I prefer straight wire whenever possible.

thanks for the insight. we do indeed have a few older keyboards in our collection. going to test this this week. Thank again.
 
Back
Top