Another plea for help diagnosing a Surf PI board

[gfcardi]

Thanks Kev! I appears that I have some work to do.

I will swap the coil leads and see what effect that has. I am troubled by the fact that you are saying that the coil is saturating. I need to make sure that there is NOTHING close to it. The "delay" pot seems to be P2, so I will adjust that to increase the time as you suggested. I will also replace the damping resistor with higher values until I get a waveform closer to the one that you provided.

Thanks for all of your help! On to the bench!

[Kev]

Ah hold off with that a moment I need to check the FETs gate voltage specs I didn't realise Q1 was sitting between 12v and -5V

[gfcardi]

Ah I see something which maybe contributing to the problem. Not all varieties of 555 have enough current sinking ability certainly not enough to charge a high side PFET arrangement like this.
Try connecting the gate of Q2 the PFET to the emitter of Q1. The bipolar variety of 555 may drive this circuit but it will use far more current.

This should turn the FET off much faster and not impede the decay. It will either way eliminate the device from the investigation.

Cheers
Kev.

Is this what you are suggesting?

[Kev]

Unfortunately that gate mod wont work unless you change R4 to 1k and then I'm not sure what it would do to your clock.
So if you've used a TLC555 timer it has a sinking value of 0.1A so it wont likely be your problem. Check the datasheet for the device you've used, because I know some can only sink a few 10s of mA

Wouldn't hurt scoping the gate of the FET and making sure you've got a nice square wave especially at TX end, then if alls good you can concentrate elsewhere.

Cheers
Kev.

[gfcardi]

OK, Thanks!

[Kev]

Oh public shame....Sorry I always get my P and N Fets mixed up, and I'm usually so obsessed with turn on times and not off which in PI is the most critical anyway.
OK switching OFF the coil we need to drive the FET high, so it's drive strength of the 555 that's important in this case not sinking, the TLC555 appears to provide 10mA.
It may need a bit of help from a pullup resistor on that gate, or the bi-polar 555 variety (i.e. NE555) which will provide up to a quarter Amp. That is off course if your waveform is drooping.
Usually you would drive it off with a transistor to ensure a very quick gate discharge and turn-off time.

Maybe someone else has spotted what I'm missing and will provide you with some real insight.

Cheers
Kev.

[Kev]

Hi gfcardi,
Is this the PCB you used?


Seems like this a workable design, people are reporting good results so I guess most of the above is barking at the wrong tree.
Sorry I stick to very narrow field in the forums, and didn't know much of the surf, but couldn't resist a plea for help.
It means then there must be a faulty component somewhere.
At least we can be pretty certain that it's before or is the preamp.

Did you follow the BOM closely as far as parts are concerned?

Cheers
Kev.

[gfcardi]

It may need a bit of help from a pullup resistor on that gate, or the bi-polar 555 variety (i.e. NE555) which will provide up to a quarter Amp. That is off course if your waveform is drooping.

The part is a NE555D, so I guess that is good.

This unit is Silverdog's "long board", which I bought as a kit (http://www.silverdog.co.uk/shop/inde...&product_id=76). Here is the BOM - partslist.txt

[gfcardi]

Some glimmer of success?

I think that I am making progress. I think that this is closer to what your describing Kev - . This was the result of increasing the damping resistor from 390 ohms to 530 ohms. Here is the result with 820 ohms - . What should the critically damped pulse look like? How do I know when it is under/over-damped?

Unfortunately I am still not able to detect anything. Any other thoughts?

Thanks for all of the help so far!!

[Kev]

Yes that looks much better, but it needs to be better still to be super sensitive. If you attach the trimmer as mentioned above, you should see the waveform begin to "ring" as you increase the resistance, you need to halt it just before ringing. But this will only apply to the coil you have on, so all other coils attached must have the exact same parameters to also be critically damped.

However I would've thought that with the improvement you have gained thus far you would've seen a correspondingly improved detection range over initial.

You're using a kit that should "work" straight off, so if you made a coil according to the specs, the 390R resistor should damp that decay curve smartly, but previously we saw it was way overloaded.
So there is definitely something fishy about the frontend, but it appears that you may well have another problem elsewhere in the circuit.
I'll go back and have a look at your scope pictures.

I would've thought that one of the other builders of this circuit would see straight away a disturbed scope shot?

Have a look at the preamp output on the scope again while you move a large metal object over the coil, and ensure that the decay curve alters accordingly.
Additionally check you preamp trimmer is set correctly and that altering it changes the baseline of you waveform. I doubt that this is the problem though since the sample gate is capturing the waveform nicely, perhaps there's an issue with the integrator or final stage. Are you able to alter the threshold to give a nice low burbling tone?

Have you extended any of the controls to pots off the board?

Cheers
Kev.

[Kev]

Hi gfcardi,
I see your scope shot of the preamp was DC so the amp offset is probably close enough.
You may need to scope the whole circuit again now that it's not so overloaded.
I'm curious to see u7 pin5 on one channel and pin13 on the other?

Cheers

[gfcardi]

Thanks!

Here is the screenshot of U7, Pins 5 & 13 -

[gfcardi]

Yes that looks much better, but it needs to be better still to be super sensitive. If you attach the trimmer as mentioned above, you should see the waveform begin to "ring" as you increase the resistance, you need to halt it just before ringing. But this will only apply to the coil you have on, so all other coils attached must have the exact same parameters to also be critically damped.

OK. I will have to dig up a pot and try what you suggested.

[gfcardi]

You're using a kit that should "work" straight off, so if you made a coil according to the specs, the 390R resistor should damp that decay curve smartly, but previously we saw it was way overloaded.
So there is definitely something fishy about the frontend, but it appears that you may well have another problem elsewhere in the circuit.
I'll go back and have a look at your scope pictures.

I would've thought that one of the other builders of this circuit would see straight away a disturbed scope shot?

The coil was purchased separately (from eBay), but it was supposedly designed for exactly this kit. I had chosen to go that route so if I did run into problems I would be certain it was not my poor workmanship building the coil.

[gfcardi]

Have a look at the preamp output on the scope again while you move a large metal object over the coil, and ensure that the decay curve alters accordingly.
Additionally check you preamp trimmer is set correctly and that altering it changes the baseline of you waveform. I doubt that this is the problem though since the sample gate is capturing the waveform nicely, perhaps there's an issue with the integrator or final stage. Are you able to alter the threshold to give a nice low burbling tone?

Have you extended any of the controls to pots off the board?

OK, this is strange...I am getting detection now! Hooray!! Yes, I can see the metal object on the preamp output. Looking good!

Yes, all controls have been extended to the offboard pots. The Threshold pot does nothing.