Modifying/upgrading a Klark Technik LB-100 active DI

[rodabod]

an earth-lift would not work since the box is phantom-powered.

Good point. Seems a shame to go to the effort of modding and essentially ending up with an opamp or two in a box though. You might be better selling and getting something decent instead.

 

[b0b]

There actually is an earth lift on the DI. There's a connection from the sleeve of the inputs to ground that can be disconnected, leaving only a small value capacitor to connect the two. I'll see if I can include it in the schematic and still be able to print it. I'm using McCad to draw my schematics. Any better free software out there for a Mac user?

 

[b0b]

Updated the schematic one more time. It now includes the inputs and the Earth Link switch. I left out the 30 dB pad on the input and 15 dB attenuator on the output because McCAD can't handle that many components.

So the verdict is that the design is poor and that's why it sounds bad? I have been trying to figure out what the  ceramic cap on the side of the transformer is doing. It seems logical to presume it's some sort of RF filter right? Measuring the transformer I get a DC resistance for the primary and secondary that's according to the data sheet. There's definitely a wire coming out of the transformer to where the right leg of the ceramic cap is connected. There is a 680R DC resistance when I connect my meter between the legs of the ceramic cap but it doesn't seem to connect anywhere else..

Well, I'd like to thank everyone who have helped me with this. It's been very informative. I guess I'd better go to the shop and get a few new DIs.

 

[abbey road d enfer]

There actually is an earth lift on the DI. There's a connection from the sleeve of the inputs to ground that can be disconnected, leaving only a small value capacitor to connect the two. I'll see if I can include it in the schematic and still be able to print it. I'm using McCad to draw my schematics. Any better free software out there for a Mac user?

Yes, the earth lift actually disconnects the jack's sleeve from the pin 1 ground. This works because of the transformer at the input, but the Pin ground is always connected to the circuit's ground, because that's the return path for the IC's power.
The ceramic cap must be in parrallels with the opamp's feedback resistor; it's a very common scheme on opamps, meant to guarantee stability (fighting oscillations). Almost any properly designed audio stage built around an opamp uses that (that's called phase lead compensation)

 

[b0b]

Ok, I have one more question... If I was to rip out the transformer and make it a proper active DI do you have any good recomendations for schematics? I found this one

here http://sound.westhost.com/project35.htm

Anyone know if this is any good? Seemes like a simple enough circuit and I should be able to modify the KT to this.

 

[AudioJohn]

As drawn the input impedance is much too low - 100K in parallel with 23K whichever input - line or speaker is selected - in fact much too low for any normal use a DI box might be employed for! This must be a mistake.....

Linked (parallel) outputs are also far from ideal - as whatever is connected to this socket further loads down the signal source, if the DI input was correct (1M ohm) - even a few metres of high capacitance cable would have an adverse effect on the signal. Together with the unknown load connected.

Next - a switch to select Phantom or battery. If the DI is left in the battery switch position - they would be run down the expensive pair of batteries.....

In a good design Phantom/battery switching should be automatic, with the battery disconnected when the input jack is withdrawn, the unbalanced output on a jack socket should be buffered and have a low impedance output, the input should be high (1M) impedance. No switches fitted - so nothing can be 'fiddled with' in use.

DI Box designs appear simple - but clearly there are many problems to address to make a successful product.

 

[abbey road d enfer]

Ok, I have one more question... If I was to rip out the transformer and make it a proper active DI do you have any good recomendations for schematics? I found this one

here http://sound.westhost.com/project35.htm

Anyone know if this is any good? Seemes like a simple enough circuit and I should be able to modify the KT to this.

As John said, there's a number of things that don't work, but they can be easily corrected.
The input pot should be increased to 1 Meg.
R1 should be several Megs (10 Megs is easy enough to find).
C1 should be a non-electrolytic. 0.1uF plastic film is fine.
Increase R2 to 1 Meg.
Install small value (100p) cap across R1.
Put a little cap (10 to 47pF) across R5.
Increase C2 & C3 to 100uF 63volt.
Increase R6 & R7 to 680R, decrease R8 & R9 to 680R.
R12 & C8 are useless, just make a direct connection from Gnd to Pin 1.
Connect negative of battery to normalling contact of Input jack (sleeve break) in order to turn off battery when instrument is disconnected.
Insert diode (1N4001...7) between positive of battery and positive of C4.

You should end up with a pretty decent box.

 

[b0b]

Thanks for that John and abbey road d enfer. Learning a lot here 

I don't have a need for the line/speaker switch or battery operation so I would leave out the two switches and D3 which according to the designer could be left out if no battery operation was needed. If there should be a small cap across R5, should there also be a small cap across R4? Aren't they setting the feedback (or something like that) for the opamp? And is R4 placed right. I thought it should be in the feedback loop for the top channel (U1A) of the TL072.

 

[b0b]

Had one last go at fixing the Klark Technik LB-100 and finally got it working. It turned out to be the transformer that was not working. I re-caped the whole board (not that many caps there) and replaced the opamp but nothing worked. The fact that the DI worked the same way as an other fully working one with the -30 dB pad in had me really confused. I went over everything with a multi-meter one more time comparing all my measurements to the same measurements on a working one and when I came to the transformer primary I found my inconsistency. The primary winding was not connecting to the pins of the transformer. A bit of extra solder fixed the problem.

A big thanks to everyone that helped me along the way.