DI needed to play guitar through consumer gear?

Hi,

I have some effects boxes that take -10dBv (I assume) a dbx 120 and a generic chinese tube-pre off ebay (its alright for a shot).
. I would like to play a guitar through them. The tube pre has 10k input Z.  I'm not sure what the inZ of the dbx is but I assume its typical hifi like 45-100k or so.

My understanding is these bits will need some form of DI conversion. Looking at the schematics of old fender amps, I seen a 1MOhm resistor before the pre-amp section,  between hot and common (IIRC).

When it comes to, commercial DI offerings they vary in price, are active passive etc but I do wonder if a passive resistor wired into some cable would suffice instead? / if those products are just after my money.

I'm earning more about how tubes work, i.e. performances curves in mA - not an expert enough to make reasoned judgment regarding this tube pre - Hell, I still find understanding in/out impedance tricky enough as it is!

Thanks for your attention I look forward to a solution

giro1991 said:

Hell, I still find understanding in/out impedance tricky enough as it is!

Click to expand...

The principal of impedance is critical if you want to do anything with electronics. Electronics is a game of impedance matching. Transistors are essentially impedance converters. Just a piece of wire has impedance. You should take the time to understand it properly. It's really not that tricky. A little abstract. But not tricky. But I suppose everything is difficult to understand until you understand it and then it's easy.

Take your guitar DI for example. Every electrical device lead has an impedance. A resistor has a very well defined impedance. It's just the value of the resistance. In the case of a guitar, it has two leads but if you connect one to earth ground, the other lead has an impedance which is defined mostly by the coil in the pickup but also by the volume and tone pots and capacitors. But with the volume and tone controls set to max the impedance is largely defined by the coil which is vaguely ~30k.

Now if you want to plug that into something that is only 10k input impedance, that's not going to work that well. At the very least your going to have a 30k/10k voltage divider so you going to loose some signal. And if you move the volume or tone controls that voltage divider is going to attenuate even more. You will also loose certain frequencies (meaning bass or treble) if there is capacitance involved, which there is.

So, as you have stated, yes, you need something with a high impedance input. Guitar inputs are usually 1M. But 100k would be fine if you don't mess with the volume or tone controls. So you asked if you could get that with a simple passive circuit? Let's see. You could have a 100k in series with 100 ohms to ground and then tap into the intersection of the two resistors. That circuit would have 100.1k input impedance which is high enough for guitar which is good and 100 ohms output impedance which is low enough for dbx or whatever so that is good. However there is an issue. A 100k/100R voltage divider is going to attenuate the signal by 60dB. Unless you're plugging into a microphone preamp (and this circuit would actually work just fine with a mic pre incidentally) the attenuation is going to be too much. Even 100k/1k is 40dB or 1/100 the signal level. So there is no way to lower the impedance and maintain the signal level. You need an amplifier.

If I were making a simple buffer amp for guitar based just on what you have described, I would use a 1M log pot in front of a non-inverting unity gain (meaning 0 gain) op amp stage followed by a 100 ohm resistor to isolate the output from capacitance and other dangers and then a large electrolytic DC blocking capacitor with a 22k resistor to ground to drain away any floating voltage. If you want to get fancy about it you might add 12dB of gain (or whatever gain makes 0 db with the knob in the 12 o'clock postion) with some feedback resistors and a large cap to ground. I would put it in a die cast aluminum box with a true bypass footswitch, power it with a 9V battery and use a low power single op amp like TL071. That would give me 1M input impedance and 100R output impedance and a gain range of -infinity to +12dB.

Hi, thanks.
So the 30/10k divider would be louder than 100.1k but it could potentially overload the dbx?

If you have an effect box with a buffered, "active" bypass you could use that in bypass mode as a driver for the 10k input. Boss pedals usually have that, Ibanez, ... most pedals with a soft switch.

giro1991 said:

Hi, thanks.
So the 30/10k divider would be louder than 100.1k but it could potentially overload the dbx?

Click to expand...

If you run the guitar with assuming ~30k source impedance directly into the dbx with 10k input, then the signal level will be about right. The dbx is probably -10dBu and the guitar can put out much more than that if you really jam on it just so. But it probably won't sound particularly great. The natural variations in source impedance of the guitar are going to be exacerbated by the low input impedance. For a rhythm part it might sound good. But for a lead guitar or picking it will sound weak. You really need a buff. Like volker said, many guitar pedals have an active bypass. Meaning when they're bypassed there's still at least a transistor follower buffering the output. That would do the trick. Most of the 80's pedals like DS-1 and so on are buffered even why bypassed.

Edit: my reply is a bit long, I guess I could simplify my concern to: can I blow anything up if I stick the gtar into either of these things? If the answer is no, and its either too quiet or too loud, then I guess I'll have somewhat narrowed down how things work.

*

lJust realised my mistake in post #1, I meant -10dBu not dBV but I see -10dbV exists aswell because people like to complicate their lives. I also hate that there is dBv and dBV. I find the old 600ohms standard mildy interesting.

Interesting that about bypass mode on pedals, I don't have any pedals, yet, haha hence these things.

"The dbx is probably -10dBu and the guitar can put out much more than that if you really jam on it just so."

Is that bad? good? I can't tell if you mean the player has to strum harder or if you mean it can break the dbx.  Is it then right to look at impedance matching as a game of voltage/level matching? (-10dbu =  0.25volts and say a guitar has 2.0V) and to just try to match these? Is that what this is? I read up on 'impedance matching' online but it was mostly concerned with cabling capacitance, the conclusion, matching (like the old 600ohm) isn't crucial, most consumer stuff is lo-z out, hi-z in sufficient to be universal. I note that tube amps (and more recently transconducance, Tz amps) worked the opposite way around. I find that interesting too, seems its largely economical / efficiency/ safety factors and cost effective driven IC solutions  that lead to that change. Shame because DOAs are bloody cool too.

Also, If active is recommended here, how are passive DI so supposed work...

Before I learned how AC and DC differed, I got in the habit of looking at signals as cars along one way road because of in/out circuit connotation but thats not how it works there isn't a IN and a OUT, its more of a MIX of capacitance, in rush. This is probably because at school here, they start on potential difference in physics without clarifying AC and DC... I excelled but then it escalated to ohms law and some other very technical stuff... I suck at math, ohms law is about as much as I could handle, they push their luck those curriculum people..

Appreciate your time anyway...  I need to practice this stuff . Any recommended reading that isn't too 'math heavy' (crap tonne equations)? I avoid it wherever possible... excluding ohms law of course, its somewhat necessary.

I'd like to learn phase, I eventually sussed with some help that its a useful 'virtual' quantity where as Group delay is the physical effect, audible, observable (GD is derivative of phase). slightly off topic but interesting. Next i'll be attempting to build a quad core processor in the garage lol but I don't have a garage and I can't so I can't.

Can you post a pic of that EBay preamp? If it’s the one with 2  6N3 tubes 4 RCA jacks and one knob there’s a schematic in a thread here. I pried off the 10k SMD resistor and just hung a 1M on a phone jack, and it had just enough gain for the line ins on a -10 board.

Also, somebody at homerecording.com posted some other hacks to increase gain.