IDX-PRO+VDI

[calebsone]

hey i am getting ready to start this project just needed some info has any thing changed from the posted info on page 7 post 316 its a file that has the hex for the vdi and the lays. if anyone knows of anything major can you let me know. and the two tone mod is that working and good to go? thanks in advance.

[Koala]

Mine test well outside the case but inconsistent depth. Others have had to add a ground to the POTs for it to run stable. When inside the case (metal) very consistent results. I get low depth in all metal mode (PP) with headphone/speakers that are too large to be driven properly. Cheaper the better I find.

Have been considering changing MPSA13 for a 14 for deep and ify signals but shallow targets are already too loud.

[Silver Dollar]

Ahh I see. That's good to know. I have a board all stuffed but haven't powered it up yet.
I had heard about grounding problems with a metal case but didn't really know what the
problem was. Thanks for the info!

[Koala]

You are right the out side track is GND. That will teach me not to use my memory getting old. It's the speaker that uses -V that's why I had to insulate it.

TX and Speaker use -V and have to be kept separate from GND

[Silver Dollar]

Hmm on the Silverdog board I see pins 3 and 12 on the lf347 tied to the surrounding etch.
That would make it chassis. Did you use a different layout?

[Koala]

Um if shield goes to gnd why do you connect the case to -V?

-V is the negative for the TX oscillator but it is still shielded with gnd and the case should also be shield...

Good question.

The design of the PCB has the -V running around the outside and the 4 mounting holes go through this. I have used a metal case. If used GND for the case I would have to drill out the mounting holes and add an insulator. Going with the -V for the case causes no problems and a factory coil runs very stable.

Only down side you have to insulate the headphone socket. Very easy to do. Just drill a big hole and use some PCB board on either side as an insulator.

[Silver Dollar]

Um if shield goes to gnd why do you connect the case to -V?

-V is the negative for the TX oscillator but it is still shielded with gnd and the case should also be shield...

[Koala]

RX ------------> GND
TX ------------> -V
Shield --------> GND

outer track of PCB = -V

I connect the -V to my case and POTs but have to insulate the headphone socket as that requires GND

[Silver Dollar]

Hmm I never noticed that -V in there. So you have RX TX Chassis and -V.
Do you use -V for TX and Chassis for RX?

[Koala]

Some people have different results but the proven method is to tie the shields
and both minus leads together inside the coil. So you only need 3 wires;

Shield/gnd
TX +
RX+

Don't think that will work on the IDX as the GND is in fact -V

fluxx both connecting like your diagram or just at the box work

Don't know which is best as I have never been 100% happy with my shield.

[Silver Dollar]

Some people have different results but the proven method is to tie the shields
and both minus leads together inside the coil. So you only need 3 wires;

Shield/gnd
TX +
RX+

[fluxx]

So, I found (on the krot2 topic) this wiring diagram. http://radiodom.ucoz.ru/_si/0/71858688.jpg Is this ok? I also found another one where the shields aren't connected to the ground up until they hit the PCB. Which one is correct? Or are they both correct? Cheers!

[fluxx]

Yes, I did the board myself. Just tested it today and found a solder bridge that I made, so that's why it's better to have professionally built PCB's. Then I found out that the 1 meg resistors were all bad (they were brand new, I've never had this happen in my entire life, but it happened). Then, the oscillator started working. With my homemade probe (DD - work in progress), the frequency is about 6.8 KHz, which I may trim down with additional capacitors to the needed frequency. Also, tried the soundcard oscilloscope and it works nicely (my oscilloscope is not available right now), and could null the probe just fine. I have not yet got to the shielding part, which from other users' experience might be a problem. I'll try and wrap it up by the end of the weekend.

As for your courageous endeavor, I applaud you. It is something I myself would try (and probably never finish, as usual), though my MCU's of choice would probably be some ARM M3's. I got used to them. Anyway, have you seen the KROT-2, here on forum. I think the guy did an excellent job with it and I will most certainly build one as soon as I'm finished with IDX. I think it's just great! He used an integrated low power power amp and also generated signal from adc. He used the same IC for speaker power as well. He has implemented a lot other functionalities (changing output frequency for different probes, etc). Perhaps use his ideas on your own design. He used an Atmel xmega, with which I have no experience, so another MCU, such as PIC might be nice.

Anyway, back to the topic. Does anyone have a wiring diagram for the DD coil that shows how it should be connected with the board. I'm mainly interested with shield wiring. Should the point of connection of the two shields be on the PCB or can I connect them together in the coil? Also, one RX wire goes to ground, should I also ground it at the PCB. This has probably been answered many times, but I couldn't find a definitive answer and there seems to be more than one approach on the subject. Are there any tested ones that just work? I'm grateful in advance for any answers I get.

[scrungy_doolittle]

Fluxx, did you do the board yourself?
I just ordered 2 idx boards and a TSGL and a hammerhead D from Andy over at silverdog.uk. One board is going to be built stock, the other is going to be a development board. I'm starting on a hybrid detector (analog and digital).
Full featured VDI is going to be the first step. I have a 6B coil on my 5900 and one on my 4900, so I want to build the idx to work with those coils.
The frequency generation is going to be done by the MCU (I'm going with a PIC24FJ128 something. Not sure whether the 44 pin will have enough pins, but I think it will.) The VDI will have enough features to let it function as a good deal more than a simple VDI, though it can be built that way. Working on pin assignments at the moment. Will support 8 bit parallel LCD display, 4 analog in's, two analog outs, 2 builtin comparators in the PIC24, and 4 control signals out for the sampling circuits. going to get rid of the ground balance and disrimination phase shifters, and just use a time delay triggered off the rising edge of the transmit frequency.

[golfnut]

Ideally it would be screened box, - because the pcb has no GND plane.