DIY DI Setup

Milothicus · #1 02-04-2009, 11:44 AM

so i noticed that Jensen sells it's DI transformers for around $75, or $80 with a kit of capacitors and resistors to make a pretty full-featured DI box...

then i was looking at the Radial JDI information, and it has a -30dB pad in addition to the -10dB and -20dB pads shown on the jensen schematic. i think the -30 dB allows it to take the output of a poweramp. is that right?

so... if i adjust the resistor values in the pad section of the jensen schematic to increase the maximum pad to -30dB, then just add switches, connectors, and a box, and you've got almost a JDI for half the price of a JDI... the only things missing are the 'merge' option, and the polarity switch, which is also fairly easily done, i would think...

has anyone successfully built one of these?

Milothicus · #2 02-04-2009, 11:47 AM

also interesting... Radial Engineering is the only Canadian company listed as a distributor for Jensen products. i've sent an email to see if they sell jensen products, or just use them...

kraigo · #3 02-04-2009, 11:52 AM

Pointer to the DI transformers please.

KO

Milothicus · #4 02-04-2009, 11:55 AM

Transformer: http://www.jensen-transformers.com/db.html

Schematic PDF: http://www.jensen-transformers.com/as/as066.pdf

Milothicus · #5 02-04-2009, 12:43 PM

or... here's a schematic that came direct from jensen for making it active:

AndyLES · #6 02-04-2009, 01:31 PM

Quote:

Originally Posted by Milothicus

i think the -30 dB allows it to take the output of a poweramp. is that right?

You still need to attach a speaker load of the correct impedance to prevent your amp from blowing up.

Milothicus · #7 02-04-2009, 01:46 PM

would it work in series with a speaker?

niftydog · #8 02-04-2009, 03:44 PM

Quote:

Originally Posted by AndyLES

You still need to attach a speaker load of the correct impedance to prevent your amp from blowing up.

That depends on the impedance the transformer presents to the amp, and what type of amp it is. Tube amps you have to take extra care with, but solid state are generally very robust.

slyjoe · #9 02-04-2009, 04:57 PM

The amp doesn't see the load of the transformer. The circuit above is going to present ~2 Megohm to the input.

niftydog · #10 02-04-2009, 06:19 PM

There's been a couple of schematics posted, that one with the FET buffer is not the relevant one to be looking at. Take a look at post no. 4.

slyjoe · #11 02-04-2009, 06:25 PM

You are correct niftydog - I skipped over that one.

Mark Reccord · #12 02-04-2009, 06:55 PM

Quote:

Originally Posted by niftydog

That depends on the impedance the transformer presents to the amp, and what type of amp it is. Tube amps you have to take extra care with, but solid state are generally very robust.

There's a resistive pad in front of the transformer for speaker level input, so the load the amp sees is effectively the top resistor in the voltage divider + the parallel combination of the bottom resistor and the primary impedance of the transformer, which is dependent on the input impedance of the mic preamp. I'd use at least 10k for the upper resistor for speaker level input. That makes for a pretty inconsequential deviation in the total load seen by the amp with a speaker in parallel with the input.

-30dB is on the edge of being enough attenuation for speaker level signals. I'd go -40, personally. 10k over 100R gives you that. Though with the JT-DB-E you're looking at 20+dB of attenuation from the transformer alone so -30 is probably fine.

That active DI posted above works but the buffer circuit leaves a lot to be desired. I built a DI with a simple jfet buffer and it was pretty high distortion. I've since designed a much better buffer circuit for my active DI. It's also easy to set the circuit up so it switches from battery to phantom by itself. I don't know why they didn't do that.

I have active and passive DIs that I designed and built myself based on the excellent Lundahl LL1530 transformer. It's not too hard to do if you have some electronics skills. They're every bit as good as the Radial or Countryman DIs.

elros · #13 02-05-2009, 03:22 AM

I was thinking of building a Jensen based DI myself..... never got around to it. Let us know how it turns out!

chicago_mike · #14 02-05-2009, 03:50 AM

Jensen transformers are usd in lots of stuff from not too expensive to sell your momma expensive. They have a plethora of schematics and many cool tube di designs too.

Milothicus · #15 02-05-2009, 10:33 AM

ok, i'm still not so clear on impedances... what sort of input impedance do i need on this DI? is the 140k enough? for some reason i see people saying i need something like 2M because i'm using a passive bass.

so... why do so many venues use standard passive DIs? i just looked up the standard one from yorkville (that's pretty much all i see around here), and it's only 50k input impedance.

i've got some fender reissue pickups in my jazz, either 60s custom shop or 62 reissues, or something along those lines... nothing out of the ordinary...

can i build this without a buffer? i'm willing to build a buffer. no problem there, but i wouldn't know how to design one.

Mark Reccord · #16 02-05-2009, 12:11 PM

Quote:

Originally Posted by Milothicus

ok, i'm still not so clear on impedances... what sort of input impedance do i need on this DI? is the 140k enough? for some reason i see people saying i need something like 2M because i'm using a passive bass.

so... why do so many venues use standard passive DIs? i just looked up the standard one from yorkville (that's pretty much all i see around here), and it's only 50k input impedance.

i've got some fender reissue pickups in my jazz, either 60s custom shop or 62 reissues, or something along those lines... nothing out of the ordinary...

can i build this without a buffer? i'm willing to build a buffer. no problem there, but i wouldn't know how to design one.

There are two issues at play if you're using a passive bass and a passive DI. Attenuation and impedance, which are inextricably linked in a transformer.

Ideally you want the DI to have an input impedance at least 10x the output impedance of the instrument. I'm not sure what the output impedance of your bass is, but passive instruments tend to range from 10k to about 50k Ohms. At 50k, you'd want about 500k at the DI input to hit that ideal 10x ratio.

The input impedance of a passive DI is dependent on the input impedance of the mic preamp it's connected to because of the way the transformer reflects the mic preamp's input impedance. The relationship is that the ratio of impedances is the square of the turns ratio. The Jensen has a turns ratio of 12:1, making the impedance ratio 144:1. Mic preamps typically have input impedances of 1k-2k. With a 1k input impedance at the mic preamp, the input impedance at the DI is 144k, which is pretty high for a passive DI, but not as high as an active buffered one. The consequence of a less that 10x impedance mismatch is that the input of the DI will load down your pickups, reducing the output signal. Some people dig this, so it's not necessarily a bad thing.

The other issue is attenuation. The Jensen provides a 12:1 voltage attenuation, which is almost -22dB. If you have a low-ish output instrument, that's going to make the mic level signal on the secondary side of the transformer very low indeed. I have found that the JDI doesn't provide enough signal at the mic preamp with certain instruments, and that's why.

Clear as mud?

My suggestion, though, is to try it fully passive at first and see if you like the results. If they're not to your liking, I can guide you through building a buffer circuit.

One advantage of a buffered input is that you can use a transformer that attenuates less than the JT-DB-E to get more output from the DI. I'm using a Lundahl 1530 in 3.5:1 mode in my active DI, which is -11dB as opposed to the -22 of the JT-DB-E.

kraigo · #17 02-07-2009, 10:54 AM

I'm curious about a passive re-amp device similar to the Little Labs Redeye. Any insights as to how to do the reverse? One article way back when (Bass Player, I think) suggested that passive DI's run backward can do the trick. Does this seem feasible? Double duty out of a passive DI? The Little Labs Redeye has a switch for doing both, so there might be a small amount more to it. The Redeye is passive, so not much more.

KO

niftydog · #18 02-08-2009, 06:16 PM

It does seem feasible to me, but I've never tried it. Perhaps the switch on the Red Eye simply enables the impedance matching circuit and/or the pad/re-amp level circuit?

kraigo · #19 02-08-2009, 07:55 PM

Quote:

Originally Posted by niftydog

It does seem feasible to me, but I've never tried it. Perhaps the switch on the Red Eye simply enables the impedance matching circuit and/or the pad/re-amp level circuit?

I would think that just running in reverse would provide all the impedance matching you'd need. A pad seems likely to me. I'm definitely not holding myself out as some sort of electronics guy though. I know just enough to not have gotten myself electrocuted yet.

KO

seanm · #20 02-08-2009, 08:10 PM

Quote:

Originally Posted by Milothicus

ok, i'm still not so clear on impedances... what sort of input impedance do i need on this DI? is the 140k enough? for some reason i see people saying i need something like 2M because i'm using a passive bass.

so... why do so many venues use standard passive DIs? i just looked up the standard one from yorkville (that's pretty much all i see around here), and it's only 50k input impedance.

I also don't know why so many people use the yorkville DI. I had one and it distorted badly with a passive bass. I think it is just that they are cheap