lobo improved

[maxbiri]

Hi Porkluvr

You are perfectly right, phase is more important respect total gain.
Because i'm not skilled about PS, i asked a friend and he suggested me to put a resistor in serie with generator, only for simulation not in real circuit, and now i have result in good agrement with yours.
I have a difference about 78 degree with and without 47nf capacitor.
If i have well understood yours plots you have about 90 degree of difference.
I think that the better solution is to leave space for capacitor on pcb and after see what happen in "real word".
The gain of rx stage is about 30 around 20kH.
I need study Lt spice because seem a very powerfull soft.

Bye Max

[porkluvr]

RX is not tuned

I think that the LOBO RX section is NOT a tuned circuit. I want to repeat one of my previous statements: phase response is a critical issue,
as is circuit gain. The receive circuit components are selected to rotate the phase so that GCX and DISC work correctly.

So, although there is a higher amplitude signal with a 47nF RX tank capacitor, it throws the RX signal completely out of whack
in relation to the TX phase reference.

You don't say which version of PS you use, but LTspice is FREE, and there is
no restriction on the number of components that you can use.

My screen shot shows your circuit (1), and the original (2). (Actually, I have modified the original circuit in #2.
There is more gain, but my modification gives almost exactly the same phase response as the original circuit
and uses components that I have on hand. More gain, but same phase response... the MC33178 @5MhZ GBW
should be able to handle it. Circuit #2 is how I will go.)

I used .0001 as coefficent of coupling between the TX and RX coils. This might represent NO target, but a small
residual backround signal - such as a slightly imperfect null or maybe a nuetral ground return.... ummh?
This "k" selection is only a guess, but:

Notice the wide variation in phase response between the two circuits. The GCX pot is set at 100% in both circuits
and (besides my modification that maintains the same phase response) the only difference is the additional 47nF
tank capacitor in circuit #1. (Not shown, there is series 51r and 11r in the TX and RX coils.)

[maxbiri]

Hi

Here are same plot about rx stage, not sure about the right way for input singnal.

[maxbiri]

Hi

I'm now working a little with ps and i began having same interesting results.
I can only simulate the input stage using mc33178, for tx stage i'm not so skilled and i cant have results.
From simulations results the best rx coil is 1.3 mH with 47nf and not 10mH so i do a big mistake thinking that values are revesed.
Tomorrow i will post plot of gain and phase with the two options.
I think that with the simulation plots we can do the right choice.
It seem that rx stage have a phase shift about -200 -220 degree.

Bye Max

[porkluvr]

see coil data chart

I notice what i think is a error, not sure, but i think that 1.3 mH must be tx coil and 10 mH the rx coil.
I'm working using a about 1.3 mH tx, maybe this is the problem i have around distorsion in tx stage?
Byr Max

That is an interesting possibility but, the 10mH TX and 1.3mH RX information agrees with the coil data chart.

http://geotech.thunting.com/cgi-bin/pages/common/index.pl?page=metdet&file=info/coils/index.dat

I have no way to know either way. What value capacitor do you use in the TX tank to resonate with 1.3mH?

I reversed the coils (to have 1.33mH + 47nF in the TX tank, 10mH+22nF RX) and adjusted the pots (giving me about 15VP-P on L1) and was able to get about 8x more output from the receiver amplifier (this is only a simulation). I have nice clean signals all around, but, there is a substantial phase offset... all other things being the same - the received signal lags an additional 10 degrees compared to using 10mH in TX.

This makes me to believe that the coil data chart is correct because otherwise the receiver section components would be different for "proper" phase alignment (as I understand "proper"). The GCX (and also DISC) would lose 10 degrees of adjustment range with a 10mH receiver coil!!!

I don't know why the TX tank capacitor's value has never been entered into any of the schematics posted on Geotech... that would be very good to know!! But I'm not complaining because there is enough to work with so that I think I can fit a LOBO into my Bounty Hunter Quick Draw II (wadda piece of junk) cabinet.

I calculate that 6.2nF resonates with 9.93mH at about 20.26kHz. (I am approximating the data for the 5X10 coil... the 8x15 coil has slightly higher inductance and resonates at about 19.7kHz). Reversing the TX and RX coils; we could use 47nF to resonate with 1.35mH (this is the medium value according to the coil chart) at about 19.96kHz.

You could use 10mH in the receiver side, but I think some RX phasing component values would best be changed in order to eliminate the 10 degrees of lag and it should not to be ignored.

[aft_72005]

Hi maxbiri
Another software which be able simulate circuit is proteus 7.
Also working whit it is simple.
Best regards.

[maxbiri]

Hi Porkluvr

Very interesting stuf!
You are perfectly right phase is more important respect simple gain but as i said i'm not skill with pspice, i need more time for remember how set spice for phase schift.
I notice what i think is a error, not sure, but i think that 1.3 mH must be tx coil and 10 mH the rx coil.
I'm working using a about 1.3 mH tx, maybe this is the problem i have around distorsion in tx stage?

Byr Max

[maxbiri]

Hi Porkluvr

Very interesting stuf!
You are perfectly right phase is more important respect simple gain but as i said i'm not skill with pspice, i need more time for remember how set spice for phase schift.
I notice what i think is a error, not sure, but i think that 1.3 mH must be tx coil and 10 mH the rx coil.
I'm working using a about 1.3 mH tx, maybe this is the problem i have around distorsion in tx stage?

Byr Max

[porkluvr]

Whoops-something i forgot

I forgot to include the actual simulation file.

[porkluvr]

I try a simulation on input stage of lobo, the 2 ls204 form a passband centered on 13.45 khz and not 20!!!!!!!!!!! the gain at 20 khz is about 36.
Bye Max

I do not have simulation models and symbols for LS204. I use the MC33178 model because I think that
LS204 and MC33178 are reasonably close substitutes for each other.
Linear Technology does NOT manufacture a close substitute for MC33178, and this "third party" model require some extra effort to get working.

LTspice was originally crated to demonstrate LTC switching regulators so third party models are added extras that must be kludged in.
By using the XOPAMP.asy symbol and editing its attributes (see the attached screenshot),
many different opamp and the LM393 comparator can be simulated. ST products (TS922 for instance) use a different
virtual pinout, and the xopamp.asy symbol must be modified to conform to this oddity... the device pins are ordered in a different sequence.
I will not go into detail... they're a pain in the neck.

The phase response of the receiver section is just as or maybe even MORE important than circuit gain. The LOBO receiver amplifier components
determine a particular phase relationship to the TX phase reference.
RX gain is only part of the problem - receiver phase is also VERY important. Receiver section components impart gain but they also cause
the necessary phase change to rotate the received signal in relation
to the TX signal so that the ground cancelling (GCX) and discriminate (DISC) pots work correctly!!!

MC33178 and LS204 opamps exhibit a slightly higher dynamic range than most garden
variety low noise & low power opamps. (Output voltage swing is better). MC33178 is superior in this respect.
Otherwise, we could simply use an LT1007 model (included with LTspice),
for instance... or LT1784 might be useful... but, these are NOT suitable substitutes, only tools for simplification
should it be necessary.

Look at my posts in "Full m/lab SD2000 schematics" - there is more information, and also some CD4000 series models and symbols.

https://www.geotech1.com/thuntings/showthread.php?t=14001&page=6

Most (but not all) of the opamp models in the opamp.sub file can be used with the xopamp.asy symbol. Xopamp+.asy is different only
because the "+" input is on top.

I know I haven't told you everything, but it's been a long day. so I'm wrapping this up. Oh yeah, notice that I have grounded the
JFET gate resistors, otherwise there would be no pull-up in this incomplete circuit.

[maxbiri]

I try a simulation on input stage of lobo, the 2 ls204 form a passband centered on 13.45 khz and not 20!!!!!!!!!!! the gain at 20 khz is about 36.
Not very nice, i will shift a little the passband centering around 20 khz.

Bye Max

[maxbiri]

Hi Porkluvr

Thanks for answer.
I forget said that every time i change coil i matched it with capacitor and setting 2 trimmer.
It's easy looking at oscope find the resonant F, it's the max amplitude point.
The strange is that when i set a coil feeding final transistor at 5 volt i have a good sinusoidal trace about 9.5 v peak- peak and a current about 2.4 mA misured in serie with R 68.
After i swich to 9 volt battery, only final transistor, and i have a bad sinusoidal trace; at this point i set the driver pot for having again a 9.5 pp undistorted signal and current misured on r 68 is now about 23-24 mA.
If i set driver pot for max tension pp i have about 48 mA and only 11-11.5 v pp.
I hope you can understand my poor english.
About PS, same years ago i used orcad ps but now i havent and i remember it was no easy to use it.
I need restart with PS, i know that it's a very powerfull soft but i need find time for restart from 0.

Bye Max

[porkluvr]

Hi all

What about????????????

Bye Max

Maxbiri,
(1) Any time you change transmit voltage supply you should adjust the drive level coming out of the 7555-pin7. Changing the emmitter resistance could also mean you should change the pin-7 pot's setting.

(2) As you change coils with different values of inductance you affect the transmitter tank circuit's natural resonant frequency; for optimum performance you should also change the parallel capacitor AND the setting of the frequency control pot near 7555-pin1 to obtain maximum output. After adjusting the frequency then you should re-check and adjust the drive level (1) for an undistorted coil voltage. These controls are interactive - you may need to do the adjustments several times. You should be aware that the frequency components and optimum drive level are not independent of each other.

I believe Tesoro makes two different coils for the LOBO but they both have very similar inductance and series resistance values. So, if you used either of the two factory coils then readjustment would not be necessary after initial setup.

Why don't you download LTspice and start getting familiar with it.

http://www.linear.com/designtools/software/

I will post a simulation circuit that you can use to visualize the information I just told you. I can't do it right now because I have something else pressing - I don't have the time - maybe tomorrow or later today. LTspice is pretty easy to use, but my circuit uses some models and symbols (potentiometer, MC33178, and LM393) that are NOT included with LTspice. These extras may need to be explained before they can be adapted into the simulation circuit.
There is LM555 included and some transistor models so you can perhaps set up the basic transmit circuit using fixed resistors -in the interim. Right click on components after they are placed and you will get the component attribute editor screen. You can enter an inductance value and add series resistance to the coil.

You would need to learn basic tasks such as placing components, wiring, rotating, mirroring and editing component value attributes. Google on "ltspice tutorial" to get some instruction, and also browse through the "Help" system.

If you simulate the circuit you can see that the potentiometer adjustments are actually quite critical; a small change in the drive level can cause substantial distortion of the phase reference voltage at the TX coil. You can also see the effects of changing frequency determining components.
Gotta go for now.

Oh I almost forgot to ask you: where are you measuring the current values that you mentioned?

[maxbiri]

Hi all

I solved problem around 7660 and started whit same test and here happen strange things!
I tested same coil i have around.
With a 2.5 mH coil i have a strong distorsion in top side of sinusoidal so a check with a 0.7 mH coil.
Drivining final transistor at 5 V i can have a very good sinusoidal shape around 20 KHz with about 9.5 V pp and 2.4 mA but when i swich at battery power (9V) i have again good shape but only 12 V pp and 48mA!!!!!!!!!!!
Increase 68 R to 100 and 150 give a little less current but signal decrease around 11 v PP.
What about????????????

Bye Max

[maxbiri]

Found 2 stupid error, one is a padstack of 78l05 with reversed pin and other is around 7660 external clock.
I will redraw schematic and pcb but this time i will post them after a good test.
Sorry for errors!

Bye Max