DI impedance tip

[mickrich]

Most DI boxes such as my BSS AR133 drop input impedance dramatically when the input pad is switched in. From 1Mohm to 47Kohm on the BSS.
This low input impedance combined with high impedance passive pickups is very bad news.

A big high frequency roll off occurs and leaves you recorded DI sounding flat and lifeless.

Your options are to leave the pad off and possibly clip your recorded DI tracks or........

Use a Boss pedal before the DI. The TU-2 is the one usually used for this job.
This is pretty well known by people in the know but in case this is news, here is how it works.
The boss TU-2 pedal acts as a buffer. It has an input impedance of 1Mohm and drops the impedance at the 2 output jacks to 1Kohm.

This is even lower than active EMG pickups which are about 10Kohm.
When the DI has a load of only 1Kohm your signal will be bright and clear and dynamic.
This is a night and day difference to plugging a passive pickup's high impedance into a too low impedance passive DI box or an active DI with pad engaged.

As an experiment I recorded my Charvel with Duncan SH-11 into my BSS with pad engaged (47K impedance) both with and without the TU-2 before the Di input.
Apart from a level 3db lower the difference in high end clarity was dramatic.
I used Logic's match EQ to get the curve then saved a matching preset with Sonnox EQ for future use when I get DI tracks affected by DI recording with too low impedance.
If your DI tracks are sounding flat and lifeless this could very well be your problem.

 

[colonel_claypoo]

thanks for sharing this information.
do you think the Korg Pitch Black tuners do the same as the Boss TU2/3?

 

[mickrich]

thanks for sharing this information.
do you think the Korg Pitch Black tuners do the same as the Boss TU2/3?

Just checked there.
Korg is true bypass so it won't buffer like the boss.
People tend to think of true bypass as a good feature and in most cases it is, but not in this role I'm afraid.

 

[bryan_kilco]

Slightly OT, but I've used my Boss NS-2 before as a signal splitter.....Out to amp, Send to interface. Worked out really well from what I recall.

 

[arvoitus]

Great information!

 

[Trevoire520]

This is a great tip Mick, I've got a AR133 and have noticed that it's possible to clip the input (admittedly only when I dig in REALLY hard on chugs) so this is a nice safeguard, cheers!

 

[JoeMeek12397]

Thanks for the tip man. Do you think a Boss DS1 would work the same?

 

[deLuther]

Guys, are you sure that simple pedal buffers at 9 V supply will not ruin your signal?
They can easily clip your signal before that will occur with DI-box. Most pedals with transistor buffer clips lower part of incoming wave.
Better to check again
If everything is good... then why you need DI-box at all, why not plug your pedal straight to line input of your interface?

There is a simple solution if you need some kind of Pad device, at least it can be simple to try, it can be built into plug or some kind adapter (from HiZ side).
For example for 1 MOhm HiZ, you can wire one small 500 kOhm resistor serial and second 1 MOhm resistor to ground (sleeve), this will give you two times less signal level and guitar "will see" the same 1 MOhm impedance. At least it should be better than plug your precious guitar into device with 9 V buffer with average opamp or transistor...

 

[mickrich]

The buffer in the TU-2 is regarded as good.
I still use a DI after the TU-2 to get a balanced signal into my interface.
Also the line input of my ISA 828 is 10k and the output from the TU-2 is 1K so 10/(10+1)x100=90% of the signal.
The BSS is 47K at -10pad so 47/(47+1)x100=98% of the signal.

10ft Elixir cable from guitar to TU-2 which sits beside the DI then a short high quality patch cable to the DI input.

I need to use the -10 pad on the BSS or hot humbuckers clip my converters even with mic pre at lowest gain. I always peak at -12dbfs when recording anything.

Suhr make a dedicated buffer box which is probably very good but using the TU-2 solves the impedance problem and sounds vastly better than not using it so it works for me. Recording DI tracks without the TU-2 in line sounds like the tone control is down on the guitar. My sonnox setting (above) works as a last resort because it brigs up hiss due to boosting the high which has been taken out, bringing up hiss as well as the guitars high end.

Another plus is I use the direct out of the TU-2 to feed the DI and the main out to send to the amp so I can mute the amp for tuning.
I use a korg rack tuner thats on an Aux from the console which I send the DI track to.

I'm not sure about impedance of other pedals but the TU-2 is the one mostly used.
You could probably google whatever pedals you have and see what the input and output impedance are. As long as the pedal is not true bypass it will buffer the signal in this way even if you have it switched off.
Also for playing live, a buffered pedal as your last pedal will send a low impedance signal to your amp. For a long (unbalanced) cable run, a low Z signal will arrive cleaner with less interference and less high freq roll off.

 

[deLuther]

Post sample of DI recorded through TU2.
TU2 can be regarded as good, but still 9V buffers can clip your signal or even worse, they can have overvoltage, in case of opamps, which leads to short-time phase reversal on opamp output when signal swing is larger than power supply.

I`m using this:

At least it as good as BSS AR-133 (compared them directly) and does not suffer from large signal because of bipolar power supply of +/-15 V (i.e. it`s like 30 V power supply)
I`m send one output to interface and second to amp for monitoring.

 

[KHE]

the buffer of TU2 definitly changes the inital waveform, it clips slightly assymetric. i have tested it by myself, if i find the files i will post them!
i built myself a DI box with a creation audio labs redeemer @ 18V and a lehle trafo for galvanic isolation if needed. best guitar-di i've ever used. (tried serveral buffer-circuits, but the redeemer sounded best. no idea whats inside)
costs were ~120$.

 

[deLuther]

the buffer of TU2 definitly changes the inital waveform, it clips slightly assymetric.

Exactly what I`m talking about.

 

[mickrich]

Slight asymmetric clipping is vastly preferable to high frequency roll off.
The DI is going to be sent through a tube screamer for re-amping which is going to clip it before it hits the amp which also clips it
EMG pickups clip the signal too, don't forget, and more than the TU-2 buffer will.
If the DI was going to be used as recorded, such as Piezo acoustic guitar or bass it might be a bit of an issue (though probably not if it's only slightly clipped)

"If it sounds good, it is good" certainly applies here in my opinion.
If you are using a passive DI (or padded active DI) with passive pickups the TU-2 buffer will seriously improve your recorded DI for either using an amp sim or re-amping.
Also as deLuther said you can also plug the low Z output of the TU-2 into a line in if the input impedance is high enough and don't need a DI box at all (at least 10 times higher impedance is preferable)
Your guitar into TU-2 into interface line in may be better than the Hi Z input on your interface in some cases.

If you have a TU-2 (not sure about other [boss] pedals) try it and see if it improves the high end clarity and dynamics of the recorded DI.

 

[deLuther]

EMGs also clips signal at transistor clipping level, but EMGs have almost symmetrical clipping, while transistor buffer clips only lower half (to some extent).
Yes, signal after screamer will be almost the same for clipped and unclipped signal, but, at least for me, better to have original clipping of pedal than serial clippind by two devices in chain.

I`m repeat about possibility of passive pad device which does not change input impedance of DI-box (active), without any modifications to DI-box itself. This can be better cure, than additional active components.

There is a simple solution if you need some kind of Pad device, at least it can be simple to try, it can be built into plug or some kind adapter (from HiZ side).
For example for 1 MOhm HiZ, you can wire one small 500 kOhm resistor serial and second 1 MOhm resistor to ground (sleeve), this will give you two times less signal level and guitar "will see" the same 1 MOhm impedance. At least it should be better than plug your precious guitar into device with 9 V buffer with average opamp or transistor...