After some hesitation, I decided to repeat a good sensor (DD METAL DETECTOR SENSOR) manufacturing technology. Since the resin option was no longer satisfactory (heavy weight, unpredictability in behavior, strong influence of temperature). Its essence consists in using polyurethane glue from Kleiberit instead of epoxy resin . Namely glue, and not various types of mounting foam. We will discuss all the pros and cons of this technology a little later. I will tell you slowly and without haste, so that nothing is missed.

First of all, look for a store that sells Kleiberit PUR-501. This is either a local or online store. The price of the glue varies quite widely: from 370 to 800 rubles and some change for a 0.5 liter container. It all depends on the greed of the seller. You shouldn’t buy a liter or more – the shelf life of this glue is only 8 months (unless you use it somewhere else), after which it will begin to thicken. But in practice, it is quite possible to store it for a couple of years (more on that later).

One DD27 size sensor will require approximately 30 ml of glue. This is approximately 30 grams in weight. Compare this with resin! The glue looks like this:

As for the preparation of the sensor case and windings, foil shielding, and only the RX coils, would be more optimal for this method. However, such shielding is not quite suitable for digital devices. Therefore, we will use a graphite screen, as the most universal, both for analog devices and for digital ones. We graphite the case. I will not describe this process. I will only say that recently I have switched to the graphite + titanium option, diluted with alcohol. It turns out very well!

Now the most important thing: it is necessary to soak the coils . We will soak them with the same titanium, strongly diluted with alcohol (without graphite). We draw the mixture into a syringe and carefully soak TX and RX. From all sides (preferably). The thing is that in the traditional version of sensor manufacturing, the turns are soaked with resin, which eliminates their displacement during operation. In this case, the glue will not soak them, so we have to resort to a trick. After applying graphite and soaking the coils, we let them lie for at least a day.

At this point, the preparatory work can be considered complete and we can proceed to laying the coils in the mold. Further, I will only cover the issues of fixing the coils and filling the sensor, and I will omit the connection and reduction.

Oh, I almost forgot, I’ll tell you a little about the weight distribution. The sensor in question was made for the Whites IDX PRO metal detector, so the coils turned out to be relatively heavy. TX – 51 g. RX – 80 g. The body itself (Bulgarian cast, internal partitions sawed off) – 104 g (together with fiberglass). The fiberglass also needs to be impregnated with diluted titanium. Here are some photos:

First, we fix the RX coil, as the most massive and voluminous. We collect 12-15 ml of glue in a 20-cc syringe. Unfortunately, I photographed the syringe from the wrong side.

We fix the case on the table and very slowly pour glue into the seat of the coil. We pour only where in the photo. The thickness of the glue is about 2-3 mm. Too much is also not desirable – otherwise it will spread over the entire sensor, and we do not need this. There is no exact recipe here – just intuitively see how much to pour. During polymerization, the glue expands. Not as much as polyurethane foam, but quite well. This must be taken into account and watch so that it does not lift the coil. It is better to immediately provide how to fix it. We poured the glue and put it in the RX groove. There is enough time for its positioning – PUR lives for 10-15 minutes.

A small addition. During polymerization, air bubbles may appear in some places (a la polyurethane foam). It’s okay – just wet the blunt end of a shish kebab stick with water and simply press them down. When the glue starts to dry, it will be elastic and you can shape it so that it completely covers the winding (although this is not essential at this stage – purely for aesthetic reasons). Leave it for about two days. A few more photos:

Let’s move on to TX. Here again, a little explanation. Many people prefer to bend or fluff the windings at their intersections (to reduce the overall height and, accordingly, slightly reduce the amount of resin). In this case, this is not required – the weight of the glue for filling the voids will be very small. So this can be neglected. The only thing is to necessarily lay pieces of fiberglass in these dreams to prevent the windings from touching.

In order for the TX winding to lie parallel to the RX plane, it is necessary to cut and glue three foam plastic bars, equal in height to the RX height. Unfortunately, I did not think to photograph the places of their gluing (on the same titanium), therefore I will show with arrows where they are. Just in case, I pulled the winding with clamps through additional foam plastic bars.

The technology is the same – pour glue into the groove (here you can use a little more than last time, the outer glued bars will not allow the glue to spread over the sensor). Then we lay the winding, position it and press it.

While the glue is hardening, use the same skewer to check for bubbles (if any). And, if desired, form the top edge of the glue layer. Set it aside for about two days.

As you can see, after all these manipulations the weight of the sensor increased by only 10 grams. At the same time, the rigidity of the structure, compared to the resin version, did not suffer at all. The windings are fixed tightly.

At the next stage, we install the cable, place the capacitors, fortunately there is enough space for them now. We assemble the entire structure together and proceed to mixing. And again a small explanation: when mixing, you should try to avoid loaded sections of the winding . In other words, there should be no internal stress in the coils. Due to the fact that the windings are impregnated, they already have an initial rigidity, which will be retained during their bending during mixing. If you slightly bend the middle part of the RX coil (for example), it will retain the adopted changes due to the previously done impregnation. Therefore, when mixing, we simply bend the coils as we need, without any fixing wedges and toothpicks. After connecting the cable, the weight increased slightly:

Now we can proceed directly to the mixing. Do not try to mix directly to zero – to avoid rephasing when bringing the sensor to the ground and changing the ambient temperature. The optimal value will be around 300 mV. In this case, a small reserve is preserved. After finishing with the mixing, smoothly move on to fixing the middle part of both coils. Here you definitely do not need to hurry, otherwise it will be very difficult or even impossible to fix something later.

A small digression . I will briefly tell you how to connect a graphite screen. We twist the braid of the conductor that goes to the cold end of the RX, fluff up the tip a little and insert it into the groove between the sensor body and the cable entry (1). We cut off a few centimeters of the MGTF mounting wire (2) and insert the stripped tip into the same groove (3) in the same way. We will then connect the graphite on the sole (or the lower layer – depending on which option you choose) to the second end. Then we fill this place with graphite + titanium. Let it settle for a day and fill it with epoxy resin + fiberglass (for strengthening). It looks like this:

Since the windings were pre-impregnated with Titan, you can use plastic clamps to facilitate their preliminary reduction. We connect everything we need and look at the oscilloscope screen. Gradually (one by one) we begin to tighten the clamps. At some point, we observe a decrease in the signal amplitude and its transition through zero. At this point, you need to tighten the clamps a little more and either leave them in this state for a couple of hours, or warm up the windings with a hair dryer. After tightening, you can remove them and begin reduction.

Oh, I almost forgot. Before mixing, you need to fix the winding intersections with resin. Well, and then gradually glue the windings from these places to the middle (with intermediate drying and balance control). Then it is guaranteed that the imbalance will not jump anywhere after the polymerization of the PUR.

Note: Don’t pay attention to the unrespectable appearance, the main thing for us is stability and reliability.

This is what it looks like:

Another small lyrical digression. While I was making the sensor, I still couldn’t decide what would be the best way to deal with the lower layer of the graphite screen. Either apply it to the sole and then suffer with its simultaneous gluing to the body, checking the quality of the connection with the upper layer along the entire perimeter of the sensor and simultaneously with positioning. Or spit on everything and apply graphite to the foam and God bless it. Yes, it’s ugly, yes, a collective farm. But it’s simpler and without a headache. In the end, I leaned towards the second option.

The sole was cut from 0.8 mm thick fiberglass. The weight was 32 grams. When I tried it on, I found a slight upward deflection of the middle part of the body. Apparently, this happened after removing the previous winding (on resin). I screwed up a little there and after polymerization the balance floated hopelessly far away. I had to use brute force to extract the guts. Now I’ll have to find a way out of the situation somehow. Sole type and weight:

Final filling before applying the bottom (or top) layer of graphite:

After applying graphite:

Now we can move on to the final part of the Marlezon ballet – gluing the protection (the sole). But before that, I needed to glue a strip of foam plastic into the middle part of the sensor (to slightly remove the deflection that I mentioned above). But that’s just me – you won’t need it.

Coat the sole with a thin layer of glue, you don’t need much. Then apply the glue to the sensor. It’s purely individual – it depends on what the distance from the graphite surface to the edges of the sensor is. I used about 10 ml, even a little less. Then place the sole on the sensor, carefully position it, press it so that it doesn’t move too much and turn it over (the sole is below, the sensor is above). Place it on a flat surface (having previously covered it with film) and adjust both parts more precisely. Then press it down with weights and leave it to dry for a day.

All that was left was to cut off the squeezed out glue and sand off the protruding edges. And I also covered the entire sensor with NC-217 varnish. It filled in the small pores and unevenness and additionally sealed the body. I plan to paint the entire structure white, but that will be a completely different story)).

Now about the most important thing: WEIGHT DISTRIBUTION. At each stage I carried out weighing. Here is what I got:

• — Body 104 gr.
• — TX winding. 51 gr.
• — RX winding. 80 gr.
• — Sole 32 gr.
• — Cable 95 gr.
• — Capacitors 6 gr.
• — Resin 4 gr.
• — Graphite 10 gr.
• — Insert (polystyrene foam) 2 g.
• – Glue 31 gr.
• — Total: 415 gr.

The balance ended up being around 440 mV.

I decided not to do a thermal shock test – winter is coming, let it rest for a couple of months. In theory, temperature fluctuations should not have a big impact on the stability of the sensor due to the low thermal conductivity of PUR. And it is also less sensitive to shocks, the sound conductivity of the bulk mass of polymerized glue is less than that of resin.

Video of testing DD27 sensor

This is where I finish my story. Maybe I missed something, but I seem to have reflected the main points and nuances. You can watch the testing video above. Well, the rest will be shown by the cop. Good luck to everyone and more loot under the sensor! With respect, DExxTER .

Source Post