Bo Hansen DI layout

[aquataur]

Greetings.
I just received my parts from Volker and stuffed them together.
I like to test my units before I plug them onto some expensive gear.

So I looked into Rod Elliott's pages for a makeshift phantom power comprised of my bench supply and two 6.8k resistors. Fig. 3.1 shows quite clearly, how this could be done. The 'scope should see the rest.

The transformer's legs should both be at the same potential, so it should not see any DC.
Do you see any hazard in this approach? Grounding through oscilloscope ground clamp?

 

[abbey road d enfer]

The transformer's legs should both be at the same potential, so it should not see any DC.

It'svery likely this DC woud end up being zero (or very close to it), so the DI would not be powered correctly. Unless your scope is floating...

Do you see any hazard in this approach?

No risk of fire or nuclear holocaust, but it may not work at all, although non-destructive).

Grounding through oscilloscope ground clamp?

Has your 'scope differential input capability? Where you use one channel as reference, with inverse polarity, and the other as test point.

I would suggest that you test the circuit, which is unbalanced and referenced to ground, and be confident in the good operation of your transformer.

 

[aquataur]

This brought up a good point. No, my scope is not floating. All scopes I encountered so far have their return line tied to earth - for reasons that I don´t understand (but never really researched too).
I always used an isolating transformer to run the scope if that was needed.

But this thing with flipping the polarity. Of course it can do that. Never new what it was good for - now I do!
Must look for a second probe.

But I just realize there is a workable solution. I report back if it works.
Edit: Rod´s circuit referenced above works as a test jig. I tied one capacitor to ground (it can have any potential, as long as the impedance is symmetric) and measured at the second cap.
There is no need to find a place for the scope return, it is internally referenced to the frequency generator.
A 1 Vp-p input yields a 0.2Vp-p output after the transformer.

 

[aquataur]

This is my rendering of the direct input box, FYI. I used a mounting method that I want to share with you.

The PCB fits more or less into a Hammond 1590 N1. I normally use 125B, but those were out of stock and the Hammond is almost exactly the same size.

I say "more or less", because although the PCB is small enough, the jacks are in the way.

I have made a mistake when drilling the holes, so don´t make the same mistake.
Drill the holes for the jacks as far to the rim (upper side when viewing from the bottom) as the connections allow for. This way the PCB can go a little further towards the jacks, and the XLR can be mounted from the inside.

Although this had been my plan in order to be able to mount all connectors from the inside, the jacks got in my way and everything had to move slightly towards the XLR, which had to be mounted from the outside. No big deal.

Note that the jacks may compete with the Molex connectors. I soldered the wires directly.
My preferred mounting is soldering nails, but the spacing of the pads is too small for using them (they shorted on their rims...). There is no real need for using exotic Molex connectors. All the more so, because all other parts are deliberately chosen to be easy to get.

I mounted the PCB onto a helper plastic plate to avoid spoiling the case with ugly mounting screws.
I used 5mm hex metal spacers with an M3 thread, mounted with a recessed screw through the plastic board. The PCB plugs into the protruding screws. The plate can be glued to the bottom with double sided tape, while the PCB can be removed if necessary.

This is quite comfortable and elegant, but raises the PCB somewhat, which fuels the space war again.
So again, it fits "almost". If somebody did their own PCB (I did not care once I realized that it fits a 125B) then this could be made smaller without any problem.

Lacking access to the more streamlined cases I made in the past, I used direct case etching.
This is always a lot of work, and the outcome is not always predictable, but I am satisfied.
Apart from that, etched cases look very professional for a hand-made unit and are (for a generic stomp box) very well readable under subdued lighting conditions. Each specimen thus becomes a peace of art, too. The shiny surface becomes dull from touch quickly, this is normal for aluminium, but it is astonishingly durable otherwise. Please forgive my amateur photography. Them shiny things are very unforgiving.

I ordered the Lundahl transformer, although I probably would not hear a difference to the other ones recommended. It sounds and works flawless upon recording, as expected.
I used the resistor values I could get from my dealer. They had a limited assortment in 0.1% resistors and I had to combine 3 values. The absolute value is irrelevant. I settled for the best match, but the 0.1% were perfectly right from the start.
Thanks Bo for the great work.

Edit: I forgot the input jack grounding when unplugged. Depicted as a red line in the guts shot.

 

[Whoops]

I used direct case etching.
This is always a lot of work, and the outcome is not always predictable, but I am satisfied.

Hi, congratulations for your work, it looks great and very professional.

Can you please describe the process of etching the case?
How is it done?

Would love to do the same for my DI builds,
thanks, best regards

 

[aquataur]

Certainly.
I whipped together a short write-up on my web site. This is far too long for here (apart from being off topic), it is just a draft, which may get extended.
But after you read about the prerequisites, you very likely don´t want to hear more about the procedure

Oh and by the way... the design of the above depicted enclosure is a hommage to the original designer. An expression of thankfulness for the huge work he has invested in this, which I am able to acknowledge.

 

[aquataur]

But after you read about the prerequisites, you very likely don´t want to hear more about the procedure

Seemingly this is what happened.

 

[guitar4444]

guitar4444,

I have measured many Lundahl transformers but not this particular one, so I don't know how it behaves in my DI-box circuit.

The LL1550 has a large core and can withstand about twice the level in the low register compared to the LL1538, and impedance/ratio configuration "C" according to the datasheet is probably the best option.

If you have access to it, try it, but I think it's unnecessary to buy one to take a chance. The LL1538/LL1538XL is hard to beat in this circuit.

--Bo

Hi Bo, Thanks very much for the reply. I have 4 of the LL1550 that I got many years ago and haven't used, so I think I will try it. I don't have much to compare it to though. Do you think it will handle as much level in the rest of the range?
thx Greg

 

[hasanova]

This was work in progress:
When I ordered the PCB's from Volker, he adviced me to buy the cheapest D.I. I could find.

This one was ideal, because it has all the hardware for re-use, like a groundpoint with screw and XLR PCB. I only needed to cut a bit on both short sides of the PCB, (to make it fit exactly between the metal sides and it already fits snug between the rubber feet) and I had to cut a bit one-side of the XLR PCB.

Not (clear) in the pic's:
- I changed the position of the red wire of the XLR to match the 1,2,3 pin connection on the main PCB.
- I added the missing 3K9 resistor and 2N5088/2N5087 transistor pair.
- I put masking tape under and along the edge of the PCB.
- just in case when one or more of the rubber feet might come of: a little piece of double-sided tape under the transformer side of the PCB, but it was already firm in place.
- I put back the SPDT switches, but haven't used them. Option for a -10dB PAD and/ or switchable impedance like mentioned previous in this thread.