Surf PI - successful build

[chrisw]

Hi George thanks for getting back to me.
So are you saying break the track coming from pin 1 and connect to pin 4. This means that Pins 1 & 4 will be connected to
the -5V.

[chrisw]

Just one point. At the moment without the capacitor connected the circuit seems to work ok. Would this not be an easier fix?

[Qiaozhi]

Hi George thanks for getting back to me.
So are you saying break the track coming from pin 1 and connect to pin 4. This means that Pins 1 & 4 will be connected to
the -5V.

Yes, that's correct.

Just one point. At the moment without the capacitor connected the circuit seems to work ok. Would this not be an easier fix?

The capacitor allows the battery voltage to rise slower than the -5V supply voltage, providing an indication at switch-on of the state of the battery charge.

[chrisw]

It still doesnt work George. I set the LED to come ar 10.5v then as I raise the voltage it stays on. I have to turn the preset
all the way round to get it to turn off. But it still does work properly if I remove the capacitor. But my circuit now is very noisy
and unstable since the last mod. I dont know if a coincidence or not or the time of night I am trying it, but before these mods the circuit seemed quite stable. I have checked and double checked to make sure I have followed your instructions.

I think I am going to leave the LED out. It seems to be giving me more trouble than its worth, and my board is starting to look messy
now.

[Qiaozhi]

It still doesnt work George. I set the LED to come ar 10.5v then as I raise the voltage it stays on. I have to turn the preset
all the way round to get it to turn off. But it still does work properly if I remove the capacitor. But my circuit now is very noisy
and unstable since the last mod. I dont know if a coincidence or not or the time of night I am trying it, but before these mods the circuit seemed quite stable. I have checked and double checked to make sure I have followed your instructions.

I think I am going to leave the LED out. It seems to be giving me more trouble than its worth, and my board is starting to look messy
now.

This might seem a silly question, but I assume you have the capacitor inserted the correct way round? Or maybe the capacitor is faulty?

[Qiaozhi]

Basically, the circuit should be working ... as long as the trimmer is connected to the inverting terminal, and the reference voltage is connected to the non-inverting terminal of the comparator. The output is open-collector, so a pull-up resistor is required, or an LED with a series resistor. If you use a normal LED, then you can leave pin 1 connected to 0V. In that case the output will switch between +5V and 0V. However, the original Crossbow used a high-voltage diode, and in that case you need to connect pin 1 to -5V, so that the output voltage will swing between +5V and -5V. At the moment (on REV-B) pin 1 is connected to 0V, which is probably the preferred option for most people.

[chrisw]

Yes I just checked it.

[chrisw]

I will wait and see if Sven has the same problem. But I think I will just leave it out of the board. Im using a 12v Lithium battery anyway not the 10x1.2v as in the original.

[chrisw]

Hey George,
What coax did you use on the original crossbow?

[Qiaozhi]

Hey George,
What coax did you use on the original crossbow?

It was nothing special: URM43 black PVC coaxial. Farnell (Element 14) 665-5858
It's basically 50ohm coax suitable for CATV/MATV, broadband, HDTV and satellite broadcasting transmission.
Probably not much good if you're trying to produce a fast coil.

[chrisw]

Thanks George.
I have two types on is a bit thinner and is steel braided and steel cored, or I have some 75 ohm which is all copper, but the overall diameter is a bit bigger.
I was wondering if I could use the steel, but I thought connecting steel to copper wasnt very good as you get some electrolytic action because of the two dissimilar metals. So I will maybe give the 75 ohm a go.

[SVEN1]

Thanks George.
I have two types on is a bit thinner and is steel braided and steel cored, or I have some 75 ohm which is all copper, but the overall diameter is a bit bigger.
I was wondering if I could use the steel, but I thought connecting steel to copper wasnt very good as you get some electrolytic action because of the two dissimilar metals. So I will maybe give the 75 ohm a go.

Is the steel coax actually steel or just tinned copper?

[chrisw]

I had scraped it to see if there was copper underneath, and I coudnt see copper. But after you posted this message I tried again and scraped harder and it is tinned copper. This also has a layer of foil as well. I dont know what the impedence is. It is a linking tv aerial cable.
So I have a choice now, Im not sure which would be better.

[chrisw]

Can I bother George again with another question;

On my crossbow board my R24 (3.3 ohm in series with the coil) is now changed to a 2W resistor, and it gets that hot I cannot keep my finger on it. Has it been designed this way or not?

I notice that the image of this resistor on the pcb looks bigger than all the other resistors on the board. So what size was it originally designed for? With all this heat I dont know if this is good to keep inside a sealed box. Or is there a problem with my build causing this?

Chris

[Qiaozhi]

There should be less than 1W dissipated in the 3R3 resistor, but has been rated at 2W for reliability reasons.
I've found my lab book containing the Crossbow notes, and here's how I calibrated the detector:

1) Adjust TX frequency (R44) to 1kHz.

2) Adjust TX pulsewidth (R34) to give a quiescent current of 200mA. i.e. Connect the coil, and put an ammeter in line with the battery during the adjustment. This should result in a TX pulsewidth of approximately 120us.

3) Disconnect coil, and set the voltage at the preamp output to 0V by adjusting the null offset pot (R2.

4) Set Sensitivity preset (R6) to mid-position.

5) Set the minimum delay setting (R18 ) to 20us (or whatever is suitable for your fastest coil) with the Delay pot at minimum.

6) Set secondary sample pulse (R19) to 400us.