Analog Target Indicator for VLF

[Nandor]

Im testing the one that Mr Gurdal shared. I know that disc pot affects VDI, was thinking about a switch for Disc and VDI select.
Finally after many trial and errors made some more progress today Im really happy now because it works much better now. just could not let it like that in the pcb's box my mind did not let me and "voila" with some logic thinking and some experimenting made this. It's still under progress.
Here is the test video with some coins and a pulltab, nickel-alu-silver.
https://www.youtube.com/watch?v=WcuZIPiQCn4

[dbanner]

Looks like it's working!! Perhaps you could try to include the meter hold and update circuit from White's 6000 patent. This might stop the meter needle from drifting after target has passed across the coil or at least the meter would hold the peak until the target crosses again. The meter moves sluggish, but perhaps it is due to it's construction.
Also, the log amplifiers, maybe they can be improved.
Nice, I like it.

[Nandor]

Yes very good point! I soldered some caps in plus for the sample and hold and a 120oHm resistor in series with needle to slow it down. Have to check that hold and update circuit thanks for pointing it out.
One very important thing is the Meter itself, it has to be big enough for all the scale and 1.5 or 2 volts. Right now the little one I have is a bit short on range and alu foil gets a really little portion as nickle also.

best regards to you!

[dbanner]

I am tempted to build a circuit. I have a big meter taken from a bounty Hunter discovery 2, it only served as intensity meter, perhaps it can serve as a vdi meter for this circuit, I'll have to check the specs on it.
You've inspired me with your success.
Regards.

[dbanner]

Looking at scheme posted by gurdal, it does have sample and hold stage for the meter. Yet your video shows the meter behavior as somewhat sluggish is the best way I can describe it.
On the 5900 patent is showing logic level pulse controlling the gate of the meter update switch.
Why not try transistor and diode arrangement for the log amps, just as shown in the patent. I don't think those transistors are anything special. The circuit might work better, who knows.
It would be nice to develop a very good working analog meter vdi circuit for home brew detector.
I'll have a go at building this circuit and experimenting.

[dbanner]

Try using a polycarbonate capacitor for sample and hold. Maybe it perform better.

[dbanner]

I forget to mention, did you have to shunt the meter?

[Nandor]

Thank you Im happy that somebody else is interested in this project. Please make your version so we can share our progress and we can improve the circuit togheter. I see great potential in this, no need for programimg and separate battery just connect it some fine tuning for each detector and go.
Have to try your ideas, I want to make another pcb. The needle moves sluggish because of the series 120ohm resistor, have not tried the shunt resistor yet.
Can you make a schematic with your ideas?

regards

[Nandor]

For the quad opamps I have used LM324 and also LF347 no difference really and for the sample and hold circuit I have tried CD4016 and CD4066 same results. I also tried FET transistors BF245 and 2N3819 instead of the CD4016 but they not really worked maybe the 2N3819. There is only 1 channel used of this IC and it would be nice to find some other purpose for CD4016 maybe switching or something else that would improve the VDI circuit.

[dbanner]

Yes, I will begin working on this project today.
My idea about the meter is to determine first by testing, what is the FSD current, check to see if there is a shunt or series resistor internal to it, or any markings that say what is the rating of the meter. For example you will see on analog multimeter face sometimes marked 50 uA 20k/VDC.
If your meter is 20,000 ohms per volt (DC), then when you work out the math - you get 50uA sensitivity. This is the sensitivity of the meter movement alone - where there is no shunt and all the current you are measuring goes through the coil.
If you want to use a 50 uA meter movement for a 1mA FSD( full scale deflection) then you can work out a shunt resistor value.
Sometimes there is a series or shunt resistor hiding inside the meter, depending upon where it came from. You can first measure how much current is require to deflect the meter full scale, then make adjustments of the shunt value to meet the criteria of your vdi circuit.
I think the White's meter is rated 1mA. So I would guess t this is the meter you want, a 1mA meter movement since the circuit you are working on is similar electrically to the Whites circuit.
If you were to take any old meter and stick in a series resistor, you won't get the proper resolution on the scale and the separation of metals won't be good on the meter even though the rest of the circuit is doing it's job well. The coil resistance also plays a part in the calculations. If we are just measuring voltage then I guess there a need for series resistor. But I am not sure. Point is we must have correct meter for the circuit.
I'm far from expert in these matters, but for circuit to work excellently and splendid, every little detail must be meticulously worked out. I can only get more knowledge and perhaps in the end, a splendid result.

[dbanner]

Did you use a pulldown resistor when you tried the FET?

[Nandor]

I think the shunt is the trimmer before the Meter itself maybe not but it sets the maximum point of it and the other 1Meg the minimum, it works pretty good. When I measured an original 6000 Di Pro SL analog meter it had 68ohm resistance if I remember well.

Man I totally forget about the pull down resistor poor FET! Have to try it again so next time maybe I can save some space and milli amps.

I've been reading about logarithmic amplifiers and the document from Analog Devices says that the input has to be a positive signal, but the differentiator filter output we use has bigger signal spikes on the negative side and those are more accentuated. Whould it be better to make a separate preamp for the VDI with non inverting inputs connected to the first filters like with the Arduino VDI?

Look at Melbeta's shematic, it's another approach. The only thing that I don't know how the Compass X80 filters are working. Eduardo shared some of he's projects with leds and even multiple sound those are interesting too. These are Patents of the IDX PRO LCD VDI, it uses comparators as phase windows and than a CD4054 drives the LED's. Fisher CZ series are with segmented VDI numbers on the Meter it only moves in 7 segments thats a nice design I really like those detectors.

I will share a Express schematic file with my ideas.

regards

[dbanner]

The White's vdi is with two log amps then a summing amp, which gives the product, but then the stage after that is doing what exactly?
I'm building the clone White's circuit with a few mods. I put a reset button included. I want to use logic level signal for sample and hold, I'm not worrying about keeping parts count low, I just want a splendid working circuit to put on IDX. I do the design on Diptrace.

[dbanner]

For better results, I try to get a hold of LF411 or similar. It has very low input offset drift, and it is better for this type of circuit(log amps).

[dbanner]

I completed a PCB layout, now I'm going to build it.
I'm using the same 5900/6000 vdi circuit with a slightly different meter update logic level drive circuit for switching the 4016 gate.
My plan is to connect to the tgsl and then tinker with the vdi circuit in order to calibrate and adjust the parameters for tgsl.
I'll use the "how hard can it be" approach.