Technical question/help

[Anonymous]

Hi all,
I have run into a problem and I'm curious if any of you might have seen this. I am using what is essentially the front end circuit Eric published here before. Looking at the signal from the coil, it looks classic. It fully recovers after about 7us. Looking at the signal on the other side of the 1K resistor that feeds the signal to the NE5534, it is spread out to about 15us and has a bit of a bend in it. I don't know how else to describe it, maybe a "knee" as the flyback is disipating. Is this normal, maybe caused by the 1N4148 diodes? Then looking at the output of the NE5534 the signal looks correct except during the flyback part of the signal there is an oscillation. This is actually where I looked first and I thought it might be ringing but looking at the signal in the earlier stages shows that it is clean with no ringing as I described above. This oscillation is occurring during the flyback and stops as the signal begins to recover which is at about 22us. I'm not sure why the signal looks so good and fully recovers from the flyback at 7us out of the coil but is now at 22us out of the front end amp as well as what is causing this oscillation. Whatever it is might also be the reason the signal is taking longer to recover after the amplifier. I tried switching out NE5534 amps to a different manufacturer with the same results. Any help or suggestions is appreciated.
Thanks,
Fred

 

[Anonymous]

I have seen virtually the same thing happen in my first prototype but without the oscillation you describe. The flyback decays in about 5 to 7 microseconds going into the 5534,, but the output is quite a bit wider. As I recall it was around 20 microseconds. The only thing that really had any effect on it for me was trying different coils and different coax to the coil which makes me think it is a combination of coil inductance and a combination of coil capacitance and cable capacitance. Also reducing the gain of the 5534 helped some. Perhaps one of the resident experts could shed some light on this problem. I am not sure if my assumptions are really correct, just my observations. Hope this helps. I just etched a board for the latest prototype that I am working on, and haven't got it finished as yet. I would really appreciate it if you could send me a copy of that front end that Eric posted. Or perhaps Eric could repost it. I did find it by doing a search in the forum, but for some reason I can't get any of the older graphics posts to load at all.
Thanks
Russ

 

[Anonymous]

Hi Fred,
What you are seeing is quite normal. Looking directly at the coil, the signal will appear faster than looking at the output of the amplifier. Low level decay signals that you cannot see directly, become readily apparent when you amplify them 400 times. Plus the 5534 is not that fast a device, so you will see its overload recovery time superimposed on that from the coil. The 1N4148 diodes are fast enough not to influence the signal. Depending on how well you wound the coil and the capacitance of the coax, you often end up with an overshoot on the 5534 output, before it settles to 0V. As to the ringing, check that you have not got any coils of wire in the vicinity. i.e reels of cable under the bench; other search coils etc. These will induce ringing from quite considerable distances.
Eric.

 

[Anonymous]

Hi Russ,
Welcome to the world of PI's. Strange things happen like those you have experienced. As you are finding out, the PI concept is simple, but the actual design or better stated, the solutions to some the problems, can be quite complex
Now, Russ has pretty well hit the nail on the head with his assessment of the problems. You need to experiment with your coil inductance, coil cable type, the cable length, the damping resistor, and even your shielding if you are trying to reduce the delay. Unfortunately, one cannot just say, do this, then do this and you will get these results. Too many things just interact.
You will see the knee in the decay that you mentioned when looking at the input of the opamp, so that is normal.
Also, there will be a delay as the signal goes through the NE 5534. If you use any compensation or if you are using a cap across the feedback resistor will most likely extend the delay through the amp a little.
You should be able to get your delay down to about 15 usec without too much difficulty. Getting it down to 12 to 13 usec is a little tough but getting the delay down to 10 usec is extremely difficult.
As for the oscillation, I would take a serious look at the power supplies. I suspect one of them is probably fluctuating. Make sure to look right at the pins of the IC. It is possible the decoupling you might be using for the NE 5534 is partially the problem. It is hard and to tell without looking at the schematic or the actual circuit in operation.
Good luck and have fun.
Reg

 

[Anonymous]

Hi Russ,
Here it is. You are right that the total recovery time is very dependant on the coil/cable capacitance and the coil inductance. Also any series resistance in the coil circuit influences it. Try an AD8055 for a much faster opamp and see the difference in total recovery time. You will need a different offset circuit to adjust for 0V on the output. There is always a trade off however; the AD8055 is rather noisy. Haven't yet found anything that is both fast and as quiet as the 5534.
Eric.

 

[Anonymous]

Eric
Thanks for the schematic. I have almost completed my latest prototype circuit board. I will try the AD8055 in the next revision. For the present I can see that I need to begin the process of optimizing the coil and cable assembly. I saw someplace a reference to some sort of insulating spacer between the coil and the shield. Do you see any advantage to doing something like that? In the few coils I have wound, I have made the shield as tight to the coil as possible.
Thanks Again
Russ_NY