OP AMPS QUESTION

[daverave]

i wonder if some of the noise could come from the 7660 power chip seeing its not in sync with the oscillator frequency....your right testing in the house is a nightmare...i must say im enjoying tinkering about with the surf pi...ive also been experimenting with the sample width and found it was about 45 Us which from what ive read maybe too wide for low conductive items so ive taken it down to abo ut 11 Us and i think there is a bit better depth on low conductive items..im going to change the 1st amp with an lt1028 when it arraives and see if it makes any difference..i can see now that the main limit on detection depth is noise.

[bklein]

Could be physical pc/breadboard layout issue? - grounding? (But seen this myself)
Would Ferrite beads on supply lines help?
Separate stages by providing dedicated battery power for testing.
Spray damping resistor with freeze spray?
I like the idea of the micro-coil. Would this be handmade or a manufactured inductor?

[daverave]

ive decided to change first amp and damping resistor to 2 watt....ive increased tx frequency much higher and seems to have slightly less noise...i think in the surf pi the pcb designer should have driven the 7660 oscillator pin from the icl 555 which i feel may have helped with some of the noise problem...

[Altra]

Hi Dave

The schematic that I have there is a 10K resistor going from the output of the 555 to the oscillator pin on the 7660

So it is synchronized, see attached

[daverave]

i have the silver dog version of the surf pi and there is no connection between the output of the tlc 555 and the 7760 oscillator pin...i may have to do a modification and make it the same as the original whites version..and see if it improves the noise level.

[bklein]

Did just changing the damping resistor make a difference?
You say that if the gain is reduced the audible noise is still there so...?
Did you try using two opamps instead of one for the preamp?

[daverave]

on the surf pi there is only one space for one 1st amp so not really an option...i guess the feedback being 1 mega ohm its going to drag in lots of noise...i have been experimenting but up until now i can still see noise when i turn up the sensivity on the scope...i have now sync'd the 7760 with the 555 timer chip which has not made much difference to the noise so i guess the noise must be very broad band external interference...i wonder if the noise could be filtered out...i have not much experience with filters but will have to experiment.

[IBGold]

You could try adding a second Opp amp in parallel piggy backed on the other one but you may have to adjust the bias to suit it works for me with Minelab SD2000.

Regards, Ian.

[daverave]

did your sd 2000 have a noise problem ??? what feedback did you use with your dual op amp ??? i know the 1st amp is the main problem where noise is concerned cause when i remove it the audio is so nice and smooth with no jitter or instability.

[daverave]

my LT 1028 op amp arraived today and tried it in the surf pi...first of all i cannot adjust the dc offset down to zero...it will only adjust down to about 0.6v...the op amp has better slew rate proformance but of course our old friend the noise is still present !!! any advice please on the DC offset problem ???

[Mechanic]

HI Dave,

Check the datasheet to see what the input path impedance should be. You might have to put a 1k resistor on the gnd side of the opamp to match the bias current. However, using a low noise opamp when you have a 1k input resistor, kind of defeats the purpose of the low noise opamp as the input path resistor noise will be dominant.
Now when you say noise, you mean audible detector false signal type noise, or noise seen on the decay curve/receive?
Is the noise present when testing away from emi sources? There will be heaps of electrical noise around your house/workshop, however this is irrelevant in the field.
I'm not all that familiar with the surf pi, so forgive me if I make some incorrect assumptions. I assume you have a 1k resistor between the opamp and coil. Disconnect this resistor from the coil side and connect it to coil gnd. Doing this will eliminate all the external emi sources and tx cct noise and allow you to hear just the noise from within the receive side of the detector.
If you still have noise present, short the output of the first amp to gnd, noise still present? Do the same to the next stage and so on until you find where the noise is at...

Cheers Mick

[daverave]

the noise is super imposed so to speak on the decay curve and im testing in a home inviroment...on the surf pi there is a resistor of 1k from coil to input of 1st amp...and a 1k resistor from inverting input to ground....if i remove 1st amp from circuit the audio is perfect smooth no jitter threshold.

[daverave]

when you say put a 1k resistor to ground of op amp...do you mean inverting input to ground or pin 4 negative supply pin ??? what resistor would you think be best from coil to pin 3 input of op amp ???

[kt315]

finding of a noise's source is not so simple issue you could expect. Foster wasted much time on that. so you see - at first, you have to find the source, at second - a method you can use againt it.
http://www.findmall.com/search.php?3...match_threads=

if power lines board tracing was projected by anyone not optimal for noise reduction you have only trouble and an impossibility to win the battle.

[Mechanic]

when you say put a 1k resistor to ground of op amp...do you mean inverting input to ground or pin 4 negative supply pin ??? what resistor would you think be best from coil to pin 3 input of op amp ???

Ahhh, see I assumed incorrectly!
Ok it has the resistor there already.

The other thing that can cause noise to pick up is your scope ground lead. The ground point should be as close to the point you are testing as possible.
If it is only visual noise on the scope, not detector response false noise, then at this point you have nothing to worry about. The filters are doing their jobs

Cheers Mick