RELIC HAWK

[nakky]

Sorry I am wrong.

[nakky]

Hi Max look at this.Grt Nakky.

[Max]

Hi Max,
All OK,
If you disconnect the 330R resistor coming to pin6 of U8 (via 470nF caps), do you still get your offset voltages? if so it may be some coupling , or that the u9 3/4/5 switch associated resitors are not doing their job ??
regards,
Fred.

Hi Fred,
must check for the 330ohm... but notice that the pin6 is dc decoupled by the 470nF cap from that elements, so I cannot see how the e.g. dc level at the 330ohm could propagate effects to that pin giving an increased offset level.

I think the offset is just by internal design on input stage of ic at U8... its own offset voltage at input... that is then amplified by gain of the stage and give the 85-90mV at pin7, causing troublesome behaviour in disc mode.

I will try to disconnect the 330ohm and see if change something or not, anyway, just to eliminate any further dubt.

Kind regards,
Max

[Fred]

Hi Max,
All OK,
If you disconnect the 330R resistor coming to pin6 of U8 (via 470nF caps), do you still get your offset voltages? if so it may be some coupling , or that the u9 3/4/5 switch associated resitors are not doing their job ??
regards,
Fred.

[Max]

Max,
I agree too that this connection doens´t make much sense, but if you connect the pot where you say the 6k8 resistor makes no sense either.

Regards,
Fred.

Hi,
yes, you read my thoughts... but problem is then, if we wanna keep output near GND level, with trimpot connected as in schematic, we need extraordinary low offset op amp... something I cannot recall about ic employed in RH... that's the tricky part of this...

The pot as in schematic could be explained just considering extremely low input offset... that could lead to wanted/expected output voltage at U8 pin7, that, from what we know, couldn't be in real , originals RHs.

How to explain the 6K8 ? I really don't know... I could explain it and existing connection of trimpot but with very "special" device at U8.

Kind regards,
Max

[Fred]

Hi Fred,
Very nice pcb you had made.How do you make it?

Hi,
I used double sided presensibilized PCB , UV and regular etching .?
But i didn´t made the clearance around non-grounded pins on the layout software, so i made this after with a drill bit.
Regards,
Fed.

[Fred]

Max,
I agree too that this connection doens´t make much sense, but if you connect the pot where you say the 6k8 resistor makes no sense either.

Regards,
Fred.

[KingJL]

.... but I think there was a mistake... read my post about.

I think you are 100 % correct about the probable mistake. The trimmer/10Meg connection makes no sense as currently drawn.

Regards,
J. L. King

[Max]

Max,
I am sure you already know this... but, the RH bandpass (-6db) in DISC mode is 3Hz to 19Hz centered at 7.5Hz. Whereas the band pass in all metal is 0 to 8Hz. In disc mode, if you do not keep the motion of the target at a fair rate the sensitivity will be poor.

Regards,
J. L. King

Hi,
you're right, but I think there was a mistake... read my post about.

Actually I get some lower detection in disc mode, probably due to the bandpass effect...

before I correct the "mistake" detection in disc mode was just 1/3 of that I have for all-metal... now difference is maximum of 5%.

Kind regards,
Max

[Max]

a mistake on schematic ?

Hi,
I think there's a mistake in schematic at U8... where trimpot connects to the 6K8 via 10megaohm resistor...

Why ? I tested the effect of turning the trimpot! In the schematic way... the effect is zero: I mean I would expect something change at output of U8 and so see something (good or bad, no matter) about audio and detection... but nothing. Trimpot seems serves nothing like connected in schematic.

Then I made a measure at pin5 of U7 when in all-metal and detector works with average results... 26cm for 1eur on air: the measure give about GND level (fluctuating a bit) at pin5 in all-metal.

Then I measured again in disc mode... and wow... I read about 85-90mV there... I mean it seems not even close to 0volts... and considering gain of U7b... well I was really astonished of that voltage there...

There's enough to saturate output stage there... without any signal...

Then I was thinking why that way of connecting the trimpot... and which kind of threshold one can achieve that way... change very few turning it when connected as in schematic, not enough to make output of U8 stay near GND level.

So, had the brilliant idea that's probably a mistake in connections and the 10megaohm must be (maybe) at pin6 of U8 to compensate stuff... et voillà... this way I can compensate to gnd as I want turning the trimpot...

I told you there must be something wrong... now I get stuff in disc mode about like in all-metal... maybe just 5% reduction of depth once tuned.

Do you think also the error was there ? Make sense to you ?

Now big troubles are about disc itself, not about detection range but about ignoring iron stuff... still doesn't work. But much probably this is pure phase issue only.

I think I will stop smoke my dang cigar for now... too much fog...

Kind regards,
Max

[zzy]

Hi,
Here is a picture of the PCB with cleared holes, i could ave done on PCB but it works that way too .
Maybe usefull if anyone want to build it.
Regards,
Fred.

Hi Fred,
Very nice pcb you had made.How do you make it?

[KingJL]

I still have troubles with disc instead, poor performance and not reliable, probably there's a mistake in schematic along disc stuff.

Max,
I am sure you already know this... but, the RH bandpass (-6db) in DISC mode is 3Hz to 19Hz centered at 7.5Hz. Whereas the band pass in all metal is 0 to 8Hz. In disc mode, if you do not keep the motion of the target at a fair rate the sensitivity will be poor.

Regards,
J. L. King

[Fred]

Hi,
Here is a picture of the PCB with cleared holes, i could ave done on PCB but it works that way too .
Maybe usefull if anyone want to build it.
Regards,
Fred.

[Max]

Dont know...maybe you should compare all three schematics. Although first two are mostly faked and inaccurate, there is chance that you can spot something and find mistake. I havent started building this yet. Sorry i cant contibute much, for now...
Regards!

P.S.
Original RH pcb is mounted in massive Al enclosure. Front and rear panels are "coated" with Al foil also. All togather (including enclosure) connected to gnd. This tells us something...?
Max try to shield your pcb in simillar manner and than check its behavior. It seems some stages are more than sensitive and easilly saturate..

Hi,
all this seems tell us, do not use it without shield!
Yes, it's really sensitive to noise... cause of its so hi amplification... many stages easy saturate with also low amount of noise... I found that even a long track on PCB make it acts like crazy...

So better would be provinding a bit of EMI shielding if possible.

Kind regards,
Max

[Fred]

Dont know...maybe you should compare all three schematics. Although first two are mostly faked and inaccurate, there is chance that you can spot something and find mistake. I havent started building this yet. Sorry i cant contibute much, for now...
Regards!

P.S.
Original RH pcb is mounted in massive Al enclosure. Front and rear panels are "coated" with Al foil also. All togather (including enclosure) connected to gnd. This tells us something...?
Max try to shield your pcb in simillar manner and than check its behavior. It seems some stages are more than sensitive and easilly saturate..

Hi Ivconic,
That is interesting,
Is this RH you have working ? if so , maybe you could take oscilloscope measures at some strategic points?
regards,
Fred.