MPP preamp is vastly improvable

[Teleno]

The square wave at the new test point #3 is wrong. Im not sure whats causing it yet but It carries through the whole integrator and is a constant tone in the speaker.

The preamp offset has to be cancelled (Q3/C6/signal "D"). It looks like it's in saturation.

You said you didn't connect C6 in your board. It's necessary. 9mV x 1000 = 9volts!

[godigit1]

Question... Is it on your positive rail as well?

Hi king JL,
had to recheck. No its not on the positive rail.
Ive done some probing around and the Problem appears at U2b at pin 5.
Hi Teleno,
I will make the changes you suggest. I have u2b wired as in the original Mpp schematic which is strange that its not working . Thats why I think I did something else wrong.

I had added C6 removed D12 the bat 54, I also seperated the power rails for the 082. The samples I had allready placed correctly I didnt even look at C and D I knew pin 10 was the EFE sample so I luckily pluged it in correctly.

My efe sample width is controled by the target sample width I would have to put in another pot to adjust it seperatly . Have to figure out where?
At 5000 pps I dont have the room for any bigger EFEb sample before the next TX cycle plus I intend to fit two more signals in there for GB. For now its working pretty good. On my handful size lump of magnets and Im getting a nice boing and signal leaving the
the coil similar to how a VLF acts on a cold rock.
I can play with that more but was also shooting to try Davores Gb soultion as well and its set up in 20 Us samples .
Im going to be going to sunnyvale for the next month for work but will take a small lab set up with me so I dont go crazy.
Thanks again for your design and helping me to trouble shooot my board.
Ill fix the Pwb once the circut is good but for now I think I can make it work.
Ill get you a picture of how I have things hooked up and my huge 56n C6 cap its kinda funny. I ordered the right value but wrong size.

[Teleno]

My efe sample width is controled by the target sample width I would have to put in another pot to adjust it seperatly.

If your EF pulse is short then you have to reduce C6 accordingly. 10n should be a ballpark figure.

At 5000 pps I dont have the room for any bigger EFEb sample before the next TX cycle plus I intend to fit two more signals in there for GB.

5000 pps seems a bit excessive to me. Indeed there's almost no time to cancel the EF before the signal decays.

For now its working pretty good. On my handful size lump of magnets and Im getting a nice boing and signal leaving the
the coil similar to how a VLF acts on a cold rock.

You lost me here, is the MPP working with the mod or not? I thought it wasn't.

[godigit1]

You lost me here, is the MPP working with the mod or not? I thought it wasn't.

I have two MPPs on the bench I was refering to the timing working ok

I set the pulse rate according to KRinAZs mods on the TX. I can definatly change things up But I figured it was at least good to get the preamp working.
Im going to go back a bit and bread board the Actual MPP preamp and integrator make sure the circut is working then make the changes .

[godigit1]

Hi Teleno,
I finally Have a place to work set up here.
I breadboarded the Mpp preamp and got it up and running.
By doing this I was able to find the mistake on my Pwb.
It was pretty obvious once I got to the J113 and went to connect to the MPP mother board. I had gate and source right on the schematic but switched on my Pwb.
So Im hoping to have your preamp running shortly.
I blew my fet the RX wire came out of the breadboard And I pulged it back in without unpowering the MPP LIttle spark and pop shes gone. Do you think I should bump it up to a irf 9640 since im replacing it or that may slow things a little on your circut?
I also made a new coil that still needs some testing but it should be pretty fast. Im going to test it as I take windings off to see where depth tops out.
Currently its at 36 wraps # 30 lits 9 strand wire 450uh and 1.9 ohms on my new round core double form 3d ss variant coil or maybe its a 6dss coil now. Im getting my 1.2 gram test nugget at 4 inches but its a little chatty . I wanted to see the transition as I take it to around 350 Uh. Theres no reason a proper 8 inch coil and set up pi shouldnt see this nugget at 8 inches at least in my mind after using a TDI and 2200v2.

Thanks

[6666]

"round core double form 3d ss variant coil"

What does it look like ?

[godigit1]

"round core double form 3d ss variant coil"

What does it look like ?

Hi 6666,
I wasnt quite ready to post on it yet I still have allot of testing to do yet . but since I mentioned it I posted a little something in the coils under new spin on a old friend.
That way I dont side track this thread .
Thanks

[bklein]

diodes question

I am a LTspice newbie playing with this circuit. I put a one ohm resistor inbetween the two diodes and the negative input of the first op amp and seem to see ~39mA.
Is this correct?
Also the circuit is intentionally running past the supply rails with the signal after that. I would think that it wouldn't be preferrable to do this as some op amps will go into phase reversal and other bad things.

[Teleno]

I am a LTspice newbie playing with this circuit. I put a one ohm resistor inbetween the two diodes and the negative input of the first op amp and seem to see ~39mA.
Is this correct?
Also the circuit is intentionally running past the supply rails with the signal after that. I would think that it wouldn't be preferrable to do this as some op amps will go into phase reversal and other bad things.
[ATTACH]45971[/ATTACH]

Remove D8, it's not needed for the high slew rate op-amps recomended.

The signal never goes past the supply rails, as far as I know.

[Qiaozhi]

I put a one ohm resistor inbetween the two diodes and the negative input of the first op amp and seem to see ~39mA.
Is this correct?

You don't need the 1 ohm resistor in order to measure current.
Use a 0V voltage source instead, which is actually a short-circuit, and will have no effect on the circuit.

[bklein]

Remove D8, it's not needed for the high slew rate op-amps recomended.

The signal never goes past the supply rails, as far as I know.

Raise your battery voltages to 10V and you will see.

[Teleno]

Raise your battery voltages to 10V and you will see.

I did. Max signal voltage is 8.1V, minimum is -8.1V.

Here's my latest version of the amplifier. Install the file LM4562.sub in directory lib/sub and the symbol LM4562.asy in directory lib/sym/Opamps.
LT1223 is a current feedback amplifier with a very high slew rate necessary to maintain the integrity of the signal at high levels of amplification.
Other current feedback amplifiers may be used: LM6171, LM6181, LM7171, AD811, AD8014, etc.

[bklein]

So that means you are hitting the supply rails on U5 and U6. U10 doesn't clip with +/-10V supplies.
Is it your intentional way of clipping the upper and lower signal portions?
Are these op amps designed/chosen to perform well doing this?

[Teleno]

So that means you are hitting the supply rails on U5 and U6. U10 doesn't clip with +/-10V supplies.
Is it your intentional way of clipping the upper and lower signal portions?
Are these op amps designed/chosen to perform well doing this?

Clipping is a different thing than having the signal going past the supply rails. What's being clipped is the signal portion dominated by the Tx coil decay, where the target signal is indistinguishable.

The opamps have been selected for their high bandwidth and slew rate to swiftly recover from saturation.

LM4562 is a 55MHz very low noise, low distortion amplifier with 20V/us slew rate. Its U1 half quickly establishes a virtual ground at the input to allow for early sampling while keeping the input noise at a low level.

LT1223 is a 100MHz, 1000V/us slew rate current feedback amplifier. Though noisy, its input noise is divided by 43 by the previous stages.

[bklein]

Ok, understood.
Back to the original question, how about the current indicated through the one ohm resistor?
What is happening there?