The Gain Change Game

[GT Blocker]

Howdy all,

I have what is a VERY newbie/stupid question. As a digital only kinda person, I'm a bit behind on the op-amp gain thing.

When looking at the Hammerhead schematic, I am not sure just which components to swap out to try different gain settings.

Let's say, for example, 500 first amp, 25 integrator and 1000 final...I'm lost. Even bumping the last stage has me wondering if I got it right.

For example:
At the moment I have replaced R29 100k with 122k and, with the new op-amps (LME49710/LM4562), things seem fine. Did I bump the gain? Did I get it right?, or is the "improvement" only an apparent one (subjective)?

The machine seems to be more sensitive, but could it just be me?

I know it's a really dumb newbie thing to ask, but dammit, this op-amp gain thing is irritating me (and I don't have time to go back to school at this age).

I hope I'm not irritating anyone or getting too pedantic, any help?
GTB

[Tinkerer]

Tweaking the HH

Howdy all,

I have what is a VERY newbie/stupid question. As a digital only kinda person, I'm a bit behind on the op-amp gain thing.

When looking at the Hammerhead schematic, I am not sure just which components to swap out to try different gain settings.

Let's say, for example, 500 first amp, 25 integrator and 1000 final...I'm lost. Even bumping the last stage has me wondering if I got it right.

For example:
At the moment I have replaced R29 100k with 122k and, with the new op-amps (LME49710/LM4562), things seem fine. Did I bump the gain? Did I get it right?, or is the "improvement" only an apparent one (subjective)?

The machine seems to be more sensitive, but could it just be me?

I know it's a really dumb newbie thing to ask, but dammit, this op-amp gain thing is irritating me (and I don't have time to go back to school at this age).

I hope I'm not irritating anyone or getting too pedantic, any help?
GTB

OK so you got your project functioning. It is now the time to do some tweaking to improve the performance.

How about increasing the sensitivity by 200%?

How about adding FE discrimination.

How can you do that?

Lets look at it. Your last schematic showed a PIC timing. Did you implement that? What is your timing schedule?
What does the output of the preamp look like?

Tinkerer

[GT Blocker]

I think you missed the Q, Tinkerer

Hey again Tinkerer,

To answer your questions:

"How about increasing the sensitivity by 200%?"
OK- How? I'm still trying to learn how to manipulate the gain at each stage.

'How about adding FE discrimination."
I know of no one who has added FE discrimination to a PI detector, so once again. How?

"Lets look at it. Your last schematic showed a PIC timing. Did you implement that?"
No. I'm still trying to adjust the gain/components on the original HHvD1 to suit the new chip set (refer to the original question at the top of the thread).

"What does the output of the preamp look like?"
Looks great! Crisp and tight. Sampling at 8uSec atm on a 9.5" concentric.

So, anyway...
Just a simple explanation of how to affect pre-amp, integrator, post-amp gains would be a good start.
I've heard thoughts that a lower pre-amp gain, low integrator gain and max out the back-end is the way to go. I've heard thoughts as to other directions to tweak gains as well.

I'd just like to understand changing gains in the HH circuit to find out for myself what works best with these two chips (not that I won't try anyway).

I'll play on copper with the pic version once I get Carl's HH version working the best I can with the new faster/quieter chip set.

Thanks again, Tinkerer
GTB

[Tinkerer]

Tweaking the HH

Hey again Tinkerer,

To answer your questions:

"How about increasing the sensitivity by 200%?"
OK- How? I'm still trying to learn how to manipulate the gain at each stage.

'How about adding FE discrimination."
I know of no one who has added FE discrimination to a PI detector, so once again. How?

"Lets look at it. Your last schematic showed a PIC timing. Did you implement that?"
No. I'm still trying to adjust the gain/components on the original HHvD1 to suit the new chip set (refer to the original question at the top of the thread).

"What does the output of the preamp look like?"
Looks great! Crisp and tight. Sampling at 8uSec atm on a 9.5" concentric.

So, anyway...
Just a simple explanation of how to affect pre-amp, integrator, post-amp gains would be a good start.
I've heard thoughts that a lower pre-amp gain, low integrator gain and max out the back-end is the way to go. I've heard thoughts as to other directions to tweak gains as well.

I'd just like to understand changing gains in the HH circuit to find out for myself what works best with these two chips (not that I won't try anyway).

I'll play on copper with the pic version once I get Carl's HH version working the best I can with the new faster/quieter chip set.

Thanks again, Tinkerer
GTB

The HH, as is, is a very good experimenter platform. It lets itself be adapted for specific jobs.
I believe your primary interest is in finding gold nuggets.
For this purpose you can use a high PPS rate and you want a short delay to the first sample.
8us sounds good. Is that with a gain of 1000? as original HH? For the HH to work well, the decay curve needs to be very close to 0V for the first sample.
The saturation of the preamp should be reduced to a minimum. One way is to reduce the preamp gain and to increase the gain further down the line to compensate.
You could try for a PPS of 4 to 5K and use a 30 Ohm resistor in series with the coil.
This is about all I can say about the standard HH.

To get more out of it, you have to make a few changes to the circuit.
If you add a FET to eliminate the flyback on the preamp, you can increase the gain on the preamp considerably.

If you change the mono coil for an IB coil, you can add a simple but quite effective FE discrimination.

This will also help a lot to reduce the power needed.

The timing will have to be changed for that.

If you are interested in experimenting, I can give you specifics for the above.

Tinkerer

[GT Blocker]

Thanks, Tinkerer

Hey there,
You are correct in that I am tailoring the machine toward small (.5-10gn) and larger gold nuggets. Since you are willing to help, let's take it a step at a time.
First the machine as it is atm:

Coil: Round concentric co-planer ~9.5" (receive coil in the center w/bucking coil). Made from 28awg 7 strand Teflon. TX = 350uH @28T / RX = 432uH @~50T Bucking coil at ~8T or so, tuning resistors for Rd - TX/RX are in the coil housing. The coil is shielded on the bottom only atm.

Pre-amp (IC6 NE5534): Replaced with the National LME49710 - Main specs: GBW= 55MHz Slew= +/- 20V/uSec Noise= 2.5 nV/root(Hz) THD= 0.00003 %
No gain mods, still the orig 1000 gain.

Integrator/post-amp (IC8a/b TL072): Replaced with the National LM4562 - Main specs: GBW= 55MHz Slew= +/- 20V/uSec Noise= 2.7 nV/root(Hz) THD= 0.00003 %
Have tried a 122k for R29, seemed to adversely affect auto tracking, though sensitivity seemed better (subjectively).
This chip set drops in seamlessly and I would like to take advantage of their specs by playing with the gains in the different stages.

So...let's start with your first suggestion:

"If you add a FET to eliminate the flyback on the preamp, you can increase the gain on the preamp considerably"

I don't understand how to do that. I thought that the dual coils did most of eliminating the flyback problem. If there's a way to tweak it a bit more with the FET idea, I'll give it a shot. But as I said, I'm not sure what you mean. I'm still working on understanding the simple gain resistors atm. For example, is the gain of the pre-amp dependent on both R13 and R16? Or just R13?

So what say we start here and then see about the integrator/post amp? If you think I can get away with it, I'll push the front end gain. What say the specs above?

I could use the help.
Thanks, Tinkerer
GTB
PS-all schematic references relate to the HHvD1

[Tinkerer]

Tweaking the HH

Hey there,
You are correct in that I am tailoring the machine toward small (.5-10gn) and larger gold nuggets. Since you are willing to help, let's take it a step at a time.
First the machine as it is atm:

Coil: Round concentric co-planer ~9.5" (receive coil in the center w/bucking coil). Made from 28awg 7 strand Teflon. TX = 350uH @28T / RX = 432uH @~50T Bucking coil at ~8T or so, tuning resistors for Rd - TX/RX are in the coil housing. The coil is shielded on the bottom only atm.

Pre-amp (IC6 NE5534): Replaced with the National LME49710 - Main specs: GBW= 55MHz Slew= +/- 20V/uSec Noise= 2.5 nV/root(Hz) THD= 0.00003 %
No gain mods, still the orig 1000 gain.

Integrator/post-amp (IC8a/b TL072): Replaced with the National LM4562 - Main specs: GBW= 55MHz Slew= +/- 20V/uSec Noise= 2.7 nV/root(Hz) THD= 0.00003 %
Have tried a 122k for R29, seemed to adversely affect auto tracking, though sensitivity seemed better (subjectively).
This chip set drops in seamlessly and I would like to take advantage of their specs by playing with the gains in the different stages.

So...let's start with your first suggestion:

"If you add a FET to eliminate the flyback on the preamp, you can increase the gain on the preamp considerably"

I don't understand how to do that. I thought that the dual coils did most of eliminating the flyback problem. If there's a way to tweak it a bit more with the FET idea, I'll give it a shot. But as I said, I'm not sure what you mean. I'm still working on understanding the simple gain resistors atm. For example, is the gain of the pre-amp dependent on both R13 and R16? Or just R13?

So what say we start here and then see about the integrator/post amp? If you think I can get away with it, I'll push the front end gain. What say the specs above?

I could use the help.
Thanks, Tinkerer
GTB
PS-all schematic references relate to the HHvD1

I am a bit busy today, but tonight I will get the HH schematic and go over it.
Probably the best way, is to analyze each building block, its functions and ways it could be modified.
I hope we can get some help from other forum members for that.

For now:
Do you use the inverting or the non inverting mode for the preamp?

You have a good choice for the coil. The Opamp specs look good too, however, each type of Opamp has its own quirks, so only testing it in the circuit will tell how it functions.

Could you post a picture of the output of the preamp?

All the best

Tinkerer

[KingJL]

I am a bit busy today, but tonight I will get the HH schematic and go over it.

Running that pre-amp with a gain of 1000 creates an effective LPF with a cutoff of 55kHz. If we want to get the best response out of these things we need to break up the preamp to 2 stages and run the gain of the first stage much lower. In the case of the pre-amp IC quoted (w/GBW 55 MHGz), the gain should ideally be <= 36.6.

Tinkerer - Get well soon. Wish you all the best!

Regards,
J. L. King

[Götz von Berlichingen]

The HH preamp looks like a Inverting Amplifier.
For Gain Calculation, see :
http://en.wikipedia.org/wiki/Operati...ting_amplifier
Am i wrong ?

[Tinkerer]

Tweaking the HH

OK, I downloaded the HH Rev D1.
I see no bucking coil. With a well balanced IB coil, things look quite different than a mono coil. Please post a scope screen shot of the TP 4 output.

For a simple explanation about IC 6:
IC 6 is in the inverting mode. If you decrease R13, you decrease the gain.
If you decrease R12, you increase the gain, but you also change the damping at the same time.
Or opposite when increasing.
Increasing the gain of IC 6 also increases the saturation of the Opamp. This increases the delay to the first sample.
When the Opamp is in saturation the gain of 1000 from 1M\1k, is not really that gain anymore.
Another important factor is the sampling point.
With a gain of 100 in the next stage, the difference between sample 1 and sample 2 will be amplified by 100. Therefore if you sample too early, when the decay curve is not close enough to 0V, the detector becomes unstable when the second stage saturates.
R 16 is a level adjust. The 5534 has provision for adjusting, however, Carl added this universal adjustment for the event when somebody changes the Opamp to a type that does not have its own adjust pins.

So there is the tinkerer's explanation.

JLKing has already given the engineer's input.

A lot more can be said. I hope a few more forum members pitch in.
If we go through the whole circuit like that we might help a lot of people.

Tinkerer

[Tinkerer]

GBW

Running that pre-amp with a gain of 1000 creates an effective LPF with a cutoff of 55kHz. If we want to get the best response out of these things we need to break up the preamp to 2 stages and run the gain of the first stage much lower. In the case of the pre-amp IC quoted (w/GBW 55 MHGz), the gain should ideally be <= 36.6.

Tinkerer - Get well soon. Wish you all the best!

Regards,
J. L. King

Thanks for the wishes. I am fine. Sent you some emails.

About the GBW. Could you expand a little on the above?

Also: a reasonably fast coil, with shielding and cable, might have a self resonant frequency of 500kHz. I look at this as an LC tank circuit that resonates at this frequency.

How can we design the preamp to make best use of this resonant frequency?

Or, what resonant frequency should we design the coil and cable assembly for?

All the best

Tinkerer

[KingJL]

Sent you some emails.

Haven't received any since 10/13.

About the GBW. Could you expand a little on the above?

GBW is usually spec'd at unity gain. As you ingrease gain the effective BW decreases (as near as I can tell, it is proportional and inverse to gain). To not extend the decay time for fast tagets (or even switching transients from blocking MOSFETs being stretched to make them apperar as targets), my experience has shown that a band pass of at least 1.5MHz provides the best results. My best results to date have been with a AD797 w/gain x20 (protected by blocking MOSFETs), followed by AD600 VGA with -6 to +73 dB dynamic range (max usable for good S/N is currently 47 dB). The no-target signal has returned to 0 baseline within 5 usec (I am currently sampling at 6 usec). The gain of the AD797 could be increased to ~x74 and still provide the 1.5MHz bandpass. With better coil design, I can improve on that.

Also: a reasonably fast coil, with shielding and cable, might have a self resonant frequency of 500kHz. I look at this as an LC tank circuit that resonates at this frequency.

My two best coils (a DD and a Fig 8: TX ~290-300uh) resonate at about 850kHz (w/shielding & coax). The Rx of both coils have a higher inductance (~420uh) and slightly extend the decay time of the flyback to >1.5usec. I plan on making a new DD with both TX and Rx <= 300uh to improve that.

To answer your question: In my opinion, as high as possible. I do everything in my power to minimize capacitance in the transmit circuit. And I watch capacitance in Rx also.

How can we design the preamp to make best use of this resonant frequency?

(1) use higher bandwidth, low noise ICs (as BW increases, I have noticed that current drain seems to somewhat follow).
(2) use dual pre-amps stages with gain reduced in each stage so as not to limit BW to < 1.5MHz.

Or, what resonant frequency should we design the coil and cable assembly for?

See above.

Note: This is all (well at least most) my opinion.... Everone has one!!

Regards
J. L. King

[GT Blocker]

ok,,,see if I get it

Hey Tinkerer and J.L. King.

Let's see if I'm starting to get it. First to Tinkerer's thoughts:

"OK, I downloaded the HH Rev D1.
I see no bucking coil. With a well balanced IB coil, things look quite different than a mono coil. Please post a scope screen shot of the TP 4 output."

So here's what the coil looks like:



Now to mix Tinkerer's thoughts with those of J.L.King:

I have kludged together the following with an LM4562 and a dip header and I think both gains are now at 33, correct?:



From the following 'schematic' (R13;HHvD1 disabled):



(Back to Tinkerer's question) the result is the following wave form at TP4 - volts/div=.2, time/div=5uSec (HHvD1):



So now my question for both of you...how can I improve this? Am I on the right track?
The wave form is cleaner (noise wise) and seems to be the same as before (except for that slight dip just before the climb back to zero, though MUCH quieter). Anyone else have any thoughts on this?

Thanks folks, GTB

[KingJL]

I have kludged together the following with an LM4562 and a dip header and I think both gains are now at 33, correct?:

Actually, no. The gain of the first stage is correct and is 33. To set the gain of the second stage configured as a non-inverting, you must also supply a gain resistor between pin 2 and ground. I would use a 1K between pin 2 of the second stage and ground. That will give you a gain of ~34 for that stage. I think ground is pin 3 of the header in your schematic.

Regards,
J. L. King

[GT Blocker]

Thanks J.L.!

Thanks for the help J.L.,
I'm a bit confused as to "I would use a 1K between pin 2 of the second stage and ground. on the second stage".
I'm assuming that you mean between Pin 6/IC1-b and ground. Since pin 6 on the amp is the inverting input of the second stage and Header Pin 3 is VB.
Of course everytime that I assume something.....
Thanks, GTB

[GT Blocker]

is this what you mean?

Hey there, J.L. King,
Is this what you meant? It's what I'm trying now and the gain is definitely better.



Would a cap across the 2nd stage gain resistors R2 or R3 help?

Thanks again, GTB