Low Noise Front End for P.I.

[KingJL]

The main reason I originally went to two opamps, instead of one, is that the scope trace showing the damping and recovery can be misleading. I found I was getting a nice clean trace, but in fact the gain of the opamp had not recovered to its full feedback bound value till rather later. I had been puzzled for a while as to why the balance of the differential integrators shifted as you went to short sample delays <15uS. This was because the opamp gain was lower in the first sample than the second. This was confirmed by injecting a 100kHz signal into the amp, along with the signal from the coil. You can clearly see the envelope narrow, as it approaches the knee of the waveform. Using the twin arrangement, I now get full gain right up to the knee.

Eric,
Could using an anolog switch or JFET switch, in front of the front end pre-amp, to block the input until just before (1-2 usec) of the first sample period help the issue. We used to use this technique in high power radar systems that had a common TX/RX path to prevent pre-amp recovery problems. Of course getting the right timing with radar was much easier, we just used the TX energy to ionize and energize the T/R protection tube.

Just a thought!!

[Reg]

Hi KingJL,

Minelab is doing something similar to what you described and it appears to work for them.

Yes, it does help to not allow the preamp to be subjected to excessive voltage and go into saturation. This problem is worse on some amps than others.

In other words, your suggestion is a good idea that can be implemented.

Reg

[JC1]

Corbyn did it

Hi all,

Corbyn did it, and the performance price is not

very high. Be sure to protect the analog switch as

shown in the schematic.

http://geotech.thunting.com/cgi-bin/...rbyn/index.dat

http://geotech.thunting.com/pages/me...byn/corbyn.pdf

Now can use any amplifier.

That should make for alot of confusion.

[Reg]

Hi JC1,

Good point!!!! I never really looked at that schematic closely to see just what all he was doing. I guess I should look more closely at the entire schematic.

Thanks again,

Reg

[JC1]

Corbyn is smart !

Hi Reg,

You might want to review

the whole article.

Magnetic Viscosity compensation

and all sorts of fun stuff. I built

something kinda like it using all different parts.

Many many years ago.

Corbyn is smart!

As are many with JC for name initials.

Coil is kinda tall and can be made DD or OO or @.

JC1

[JC1]

MAX352

Hi All,

I use Maxim MAX352 analog switches.

Lot better than that 4053 or other 4000 series stuff.

Little more pricey but hey, the first find makes it worth it.

[JC1]

specs

Max352 analog switches

* Low On Resistance < 22Ω Typical (35Ω Max)
* Guaranteed Matched On Resistance Between Channels < 2Ω
* Guaranteed Flat On Resistance Over Analog Signal Range Δ3Ω Max
* Guaranteed Charge Injection < 10pC
* Guaranteed Off-Channel Leakage < 6nA at +85°C
* ESD Guaranteed > 2000V per Method 3015.7
* Single-Supply Operation (+10V to +30V) Bipolar-Supply Operation (±4.5V to ±20V)
* TTL-/CMOS-Logic Compatibility
* Rail-to-Rail Analog Signal Handling Capability

http://datasheets.maxim-ic.com/en/ds/MAX351-MAX353.pdf


I got a bunch of them used for years.

There may be something better but these aren't bad.

[bugwhiskers]

Would the AD 8099 be of any use?
see:http://www.analog.com/UploadedFiles/...34AD8099_b.pdf
ElectroNovice

Hi Electronovice,

The LF noise figures on page 13 are fairly typical in that they start to go through the roof below 10 Hz. The AD797 is far better in that region.

regards
bugwhiskers

[Unregistered]

Low noise

Hi Treasurediver,

The AD797 certainly looks like a good one to try. I'll see if I can get one.

Thanks, Eric.

pike
use 100ohm feed back resistors and tandem 3 opamps in parallel. 2 watt low ohm ressistors give low noise

[Unregistered]

Low noise

Hi Treasurediver,

The AD797 certainly looks like a good one to try. I'll see if I can get one.

Thanks, Eric.

pike
use 100ohm feed back metal film resistors and tandem 3 opamps in parallel. 2 watt low ohm ressistors give low noise

[Unregistered]

ad797

can u send us a schematic of what you are saying

[GT Blocker]

gonna try a weird one...

I ordered a few Nat. Semi. LME49710's for a try on the front end amp...I'll let every one know how they do in a few wks or so (UPS lost my shipment for most of my HH project, so I'm a bit behind as I wait). The specs seem good...any thoughts, all....

PS! **** Does anyone have a line on spray/liquid positive etch resist? It seems that GC no longer makes it.

[Max]

I ordered a few Nat. Semi. LME49710's for a try on the front end amp...I'll let every one know how they do in a few wks or so (UPS lost my shipment for most of my HH project, so I'm a bit behind as I wait). The specs seem good...any thoughts, all....

PS! **** Does anyone have a line on spray/liquid positive etch resist? It seems that GC no longer makes it.

Hi,
you can use "positiv 20" , can find on ebay and many dealers still have it today.

Like this:
http://www.rapidonline.com/Tools-Fas...o-resist/30091

Though requires more care than other I've used in past it's a good product if you follow process step by step. Very important is drying (baking process) in an oven... 80°C for 20 minutes will be fine in most of the cases.

I've used also other products... some made in France , good quality but they require lot of care about layer thickness, so better positiv20.

Kind regards,
Max

[Unregistered]

Coil and Noise in PI's

Hi,
Kind regards,
Max

Pike here
Whats the shell made of for the coil housing. Some plastic housing absorb the Magnetic fields reducing the intensity to half. What about melaware, fibre (dining table food eating plates) for search head housing.

SEAN GODDARD has some best designs

We are measuring few micro volts and Noise has to be dealt carefully, the 1st stage opmp after coil go unstable with 1meg feed back resistor (use 3 opapms in tandem and use 330ohm feed back resistor), also the noise induced is very high depends indirectly to size of resistor. Also the coil damping resistor should be metal film and 2watts is better for PI's,take the breakdown voltage for resistor as 200V, so if a coil gives 300V spike the the resistor breaks, use 2-3 resistors in parallel ....there are other issues too which mos people and professionals dont consider and design and type here in an unprofessional way...

[Max]

Pike here
Whats the shell made of for the coil housing. Some plastic housing absorb the Magnetic fields reducing the intensity to half. What about melaware, fibre (dining table food eating plates) for search head housing.

SEAN GODDARD has some best designs

We are measuring few micro volts and Noise has to be dealt carefully, the 1st stage opmp after coil go unstable with 1meg feed back resistor (use 3 opapms in tandem and use 330ohm feed back resistor), also the noise induced is very high depends indirectly to size of resistor. Also the coil damping resistor should be metal film and 2watts is better for PI's,take the breakdown voltage for resistor as 200V, so if a coil gives 300V spike the the resistor breaks, use 2-3 resistors in parallel ....there are other issues too which mos people and professionals dont consider and design and type here in an unprofessional way...

Still pike here ?

So take this...

I don't understand why you guys have that battles with PIs that are well known to be so bad on real inland conditions... this surprise me much more than 3 op amps and resistance splitting!

Kind regards,
Max