RELIC HAWK

[Max]

Tested an NPN with hfe=600 there and change nothing... just lower resistance coil works!

Neverending story with this oscillator...

[Max]

Hi,
I think I found the source of all troubles... at least in my previous simulation... the dang 22K resistor I have in the circuit I had like 27k in previous simulation.

Now, I changed it to 22K and got 13.6Vpp for a 2.5ohm resistance coil, 1mH and about 18Vpp if the coil resistance drops to 2.2ohm.

So, for the simulator... the 2.5ohm coil MUST work though not optimal one... and with ideally maximum voltage of 13.6Vpp on it.

Now the problem is that I have the dang 22K in the real pcb but it doesn't work... cause probably the NPN I've used is low gain BUT I used an BC547B so still don't understand that, must check again.

Simulations are cool... but often reality is truly different stuff..

Kind regards,
Max

[Max]

Hello, I measure resistance with multimeter PC5000 of the company SANWA. Resistance of the cable-probes when they are in short-circuit was 0.08ohm.
Best regards

Hi,
this is an accurate one: 0.2%rdg+ 6dgt on 500ohm endscale.

So I think your measure is probably one of the most accurate we have on that coil.

We have to understand why works on its circuit and not on the cloned one...

Kind regards,
Max

[Max]

Max, I think the transistor that will have the greatest impact will be the NPN that is driving the PNP cascade and provides the oscillator feedback.

Regards,
J. L. King

Hi,
I agree and will make some tests about. I also spotted an interesting stuff when testing with a power supply with current limiter... sometimes also with the 1mm wire coil it fails to start.

When you use a 12v battery this never happens and the low resistance coil works on that dang oscillator... but using limit on current at supply give so drop the voltage that it doesn't start. If I rise the current to 500mA limit I can start oscillator each time but at e.g. 200mA it doesn't start at all.

I retested the 0.5mm coil but without success so far... even using a 12v gel type accumulator.

I will try different NPN there to see what happens. Tested also tl074 model on simulator but all the same story so op. amp. impact is few unless of using very special slew rate parts.

Kind regards,
Max

[KingJL]

Hi Eugene,
the BC557 and BC558 are both PNP and about same stuff so not a problem changing one for another in most of the cases.

I suppose the real original RH has different ones inside... that's the real trouble... I think I will simulate a few well know types and see if can sustain oscillations at 2.5ohm or higher resistance.

Max, I think the transistor that will have the greatest impact will be the NPN that is driving the PNP cascade and provides the oscillator feedback.

Regards,
J. L. King

With 850uH , 1mm wire coil I get it oscillate pretty easy, with strong magnetic field around.

Kind regards,
Max

[VUJO]

Hello, I measure resistance with multimeter PC5000 of the company SANWA. Resistance of the cable-probes when they are in short-circuit was 0.08ohm.
Best regards

[Max]

Hello Max, Fred , Everyone , I was just wondering is the transistor BC-557 next to tx-coil correct which would be NPN. On the copied Cheap-Version of Relic-Hawk they are using BC-558 a PNP !!! Bad oscillation occuring at tx-coil because of this ? OR ,If Rev. 1.6 values are correct , then dc resistance , Coil wire gauge and number of windings are not o.k.??.............Thanks....Eugene

Hi Eugene,
the BC557 and BC558 are both PNP and about same stuff so not a problem changing one for another in most of the cases.

I suppose the real original RH has different ones inside... that's the real trouble... I think I will simulate a few well know types and see if can sustain oscillations at 2.5ohm or higher resistance.

With 850uH , 1mm wire coil I get it oscillate pretty easy, with strong magnetic field around.

Kind regards,
Max

[Max]

Hello Max and other friends on the best forum. Regarding problems with Tx coil I have some other experience. I have original PCB of COINSEEKERP2Pro and appropriate coil DD (D = 265mm).Device has a good discrimination metals, but very little of depth. Coil Tx has 3.75ohm resistance and oscillator for a stable 4875Hz. But how much windings coil has and what is his diameter wire I don’t know. Beside that I tried on proto board to put the same configuration oscillator with the same components and with coil of COINSEEKERP2Pro, and it did not work.
Best regards
Vujo

Hi,
thanks for informations. Who did you measure that value ? Multimeter ?

I ask cause many (also good ones) multimeters have an offset of 1-2ohm about usually.

If so... and you read 3.5 ohm there was probably 1.5 or 2.5 ohm instead , or anything between.

In severe cases the offset could be about same extent or more than real resistance being measures... so you read 3.5 but are 0.5 or 0.4 there.

I suppose the standard ML coil there is very low on resistance, but cause we don't know components used in original RH we can only guess... so is possible also that coil is really 3.5 or 2.5 but works cause of components used.

Kind regards,
Max

[Max]

Max,
Just to test out the theory that the op-amp has considerable impact, I substututed the op-amp for one with a very high slew rate (LT1814 quad opamp, Slew Rate = 750V/usec). Using an 8.2V zener for D1, I can use a coil with up to 3.4 ohms series resistance.

Regards,
J. L. King

Hi,
yes you're right. All components there are critical. I think the very bad story is about transistors... cause I truly think in original RH they couldn't be BC557 or BC547 ("european" kind) but some exotic transistor type... maybe also selected about hfe and similar. About IC same stuff: probably original have nothing to do with TL0xx series but it was maybe an TS27M4 or similar stuff: BTW I tested the 27M4 there and change nothing so other stuff is involved sure.

I strongly think coil resistance is a key factor too... and if wire diameter is something 0.7 or 0.8mm it's clear that the lowering of resistance will made easy the oscillator start running properly.

I think, instead, zener value is like in original cause it's easy/cheap stuff and people who made schematic of e.g. coinseeker put it there unmodified.

Kind regards,
Max

[VUJO]

Hello Max and other friends on the best forum. Regarding problems with Tx coil I have some other experience. I have original PCB of COINSEEKERP2Pro and appropriate coil DD (D = 265mm).Device has a good discrimination metals, but very little of depth. Coil Tx has 3.75ohm resistance and oscillator for a stable 4875Hz. But how much windings coil has and what is his diameter wire I don’t know. Beside that I tried on proto board to put the same configuration oscillator with the same components and with coil of COINSEEKERP2Pro, and it did not work.
Best regards
Vujo

[KingJL]

Max,
Just to test out the theory that the op-amp has considerable impact, I substututed the op-amp for one with a very high slew rate (LT1814 quad opamp, Slew Rate = 750V/usec). Using an 8.2V zener for D1, I can use a coil with up to 3.4 ohms series resistance.

Regards,
J. L. King

[KingJL]

Max,
The best stability seems to be obtained when the zener diode, connected to the base of the BC577C (the one that you had as a standard diode), is a 8.2V component. This does not solve the problem, just extends the threshold slightly.

It appears that we do not have sufficient feedback through the NPN to sustain oscillations with coil series resistance above 2.5ohm. That is why the op-amp type has effect. I think all of the oscillator components; coil, transistors, and op-amp are critical.

Regards,
J. L. King

[amtech2005]

RELIC HAWK

Hello Max, Fred , Everyone , I was just wondering is the transistor BC-557 next to tx-coil correct which would be NPN. On the copied Cheap-Version of Relic-Hawk they are using BC-558 a PNP !!! Bad oscillation occuring at tx-coil because of this ? OR ,If Rev. 1.6 values are correct , then dc resistance , Coil wire gauge and number of windings are not o.k.??.............Thanks....Eugene

[Fred]

What to say... I must find some wire and wound again with thicker copper. I think also this is a case where aluminium wire, also square shaped, could be of help... but where to find some ?

Kind regards,
Max

Hi Max,
There is sometimes aluminium wire in he degauss coil of TV sets,not all but some.Old Philips ones for example.
Regards,
Fred.

[Max]

I agree about coil being critical component. My simulations show that after 2.5 ohms, the coil current starts to decrease rapidly. Anything over 3 ohms stops oscillations. This is one coil where Litz wire may be very beneficial.

The op-amp does have an effect (probably due to the slew rates), although I agree that it will not fix it. The TL074 so far works best in my simulations. All others have a slower oscillator startup or less coil current.

Regards,
J. L. King

Hi,
I think a good solution could be using thicker wire... anything from 0.6mm to 0.85mm will be nice. The coil I've tested so far was made of 1mm copper wire!

but 1mm is too much for my taste... heavy copper and also too thick wound I don't like at all.

Litz could be an alternative... but the problem is finding so large litz wire... not so easy for me, I have the small kind but not thick ones.

I'm still wondering why I get so bad results with 0.5mm wire... but simulations show clear the effect of too resistance there, so there's nothing bad about 0.5mm but the fact oscillator stage is very critical this way.

I'm also wondering of Coinseeker schematics and data: use TL082... I've used TL084 for tests (and also LS404) and was a disaster!
With TL074 change nothing, same stuff... no oscillations on 0.5mm wire coil.

What to say... I must find some wire and wound again with thicker copper. I think also this is a case where aluminium wire, also square shaped, could be of help... but where to find some ?

Kind regards,
Max