Hammerhead info

[Reg]

Hi All,

I finally got around to building up a HH PI. Initially, the decay curve was such that I couldn't use a delay much below 16 usec or so. I might have been able to reduce it some by adjusting my damping resistor which was at 800 ohms, but decided to try something different. I had selected an 800 ohm resistor in hopes that my coils built for my GQ clone would work.

Made a simple mod and changed R12 from a 1K to a 2.2K. Now I was able to use the minimum delay which was about 10 usec. I was using coils I built for my modified GQ clone and they worked about as well on this PI also. I forgot to mention that I am not using the 740 FET, but am using a FQPF 2N 30. this FET has an on resistance of about 3 ohms or so. What caused me to select this FET was the low capacitance values it displayed.

I also modified the HH so I could use a DD coil. This is a very simple thing to do.

I wasn't satisfied with the sensitivity of the HH to small gold so I increased the sensitivity of the diff amp and the autotune amp. I also expanded the range of the frequency by changing the 100K to a 250K pot.

Sample width was tried at various widths and I finally left it about 10 usec. I might reduce this some, but not much. Reducing it much will reduce the gain. Increasing the sample width seemed to increase the noise level.

So far, the HH doesn't perform as well as my modified GQ clone, but it does quite well. Detection of a 5 grain nugget is about 3" using a 11" DD coil. A nickel range was a little over a foot.

This isn't bad for a home built PI project. It is not complete yet, by so far all has worked out quite well. I would like to change the audio though. Still giving that some thought.


Reg

[Beachcomber]

Re: Hammerhead info

Very Good. Keep us updated on your mods and how they work out. I am planning on building two Hammerheads over the Summer...once I finish the projects that I am currently working on. I am going to build the first HH on the large PC board and once I have it working the way I want I am going to build the SMD version incorporating the changes so my final version will be SMD. I have both the first full size board that Carl made and the newer board and I am going to order the smd boards from him this week. It should be fun and challenging to build the SMD version....though SMD stuff doesn't take well to experimentation. It's a more of a build it once and do it right the first time type kind of thing

Thanks Again for you Post!!!

HH

Beachcomber

[Reg]

Re: OOPS

Hi All,

I mentioned I was using the FQPF 2N 30 for the FET. Well, I still am but ran into something I didn't realize.

The general specs on the parametric search indicate the above FET is rated at a 2.1A continouus current. Since the pulse time is extremely short when compared to the cycle time, the current is no where near the point of being a problem.

Well, going back over the specs of this FET and others, I noticed that the actual datasheets rate this FET at 1.34A continuous, and 5.36A pulsed. Well, this is a little lower than I like but it still should be safe. One could just use a heatsink and keep the pulse on time short as an added safety measure if this FET was purchased. I personally will not even worry about it at this time on my PI.

One could also use the FQI2n30 which has the same resistance and has a rating of 2.1A on both the parametric search and the data sheets. It also has very low capacitance values, like the one I am using.

BTW, the resistance of my coil is about 3.5 ohms, the FET is about 3.7 ohms, and the 1 ohm series resistor, R10, adds up to 8.2 ohms. The result will be peak pulse of about 1.5 A with 12V applied. This results in an average current of about 150 ma or so. Actually, the range of current I see at the ps leads is a max of about .210A and reduces to less than 50 ma through the range of the frequency and pulse width control.

The actual current will depend upon the pulse width and frequency, so it can vary. Since my objective is to see how well it will work for hunting gold, a short pulse works fine.

Some of you might say that the 1 ohm shouldn't be added into the above equation to determine the current because of C 10. Well, I moved the C10 connection to the other side of R 10 so it is effectively in parallel with C1. However, it is located very close to the FET which helps.

BTW, after shielding the case and making a few more changes, the PI does work quite well.

It looks like AA Ni-Mh's should work just fine. I would use 10 of them, though.

Presently I am using some surplus batteries I got from BG Micro. I am not sure how many they have left but for those interested, they have 8.4V 3800 ma packs (pt # BAT1060) for $6.95. The packs are built up with 7 4/3A Sanyo batteries, which are quite a bit bigger than AA's. One just has to use 2 packs and remove the extra batteries to obtain the desired voltage. I use 10 batteries. The nice thing is, these batteries have plenty of "oomph" for those who want to build a PI that pulses with much more current. As for me, I do not see any advantage to increasing the current at this time.

Obviously, if a lot more pulse current is desired, then a different FET should be chosen.

Reg



Battery packs