Need Guru to look at 5534 scope shot,something is wrong.

[Brian Deese]

Here is a video I made this afternoon,inside my house. https://www.youtube.com/watch?v=w_35...ature=youtu.be

[Brian Deese]

I fixed the broken unit about two weeks ago,the overvoltage protection diode on the vc supply was shorted.I removed it and it works just fine,just like the other one. I looked at it on the scope and it is the same in all respects. So it was "as designed" operstion after all. So I have been learning,and collecting information on this great forum for some time now. I decided since I have two of these that I will mod it/them. So for the last two weeks I have been at it with the soldering iron. I have had some good luck with it,and I'm very pleased with it. I got my head wrapped around a couple of circuits that I didn't understand previously. I am uploading a video to youtube right now,I'll post the link when its finished. I will also post the mods that I did to it,and the gains/losses from those mods.

[Brian Deese]

Well I have been away working forever,I wandered into the local army surplus store down here in Wilmington. There was a vintage AF-108 laying there,and it was cheap. So I bought it,thinking I will compare the two machines and see if they are the same.However the new machine is faulty,Q2 the voltage reg transistor gets so hot you cant touch it in about 60 seconds,no output,no lights.I will have to get it home and check it,hopefully Q2 which is a buy92 is just shorted. How do pnp transistors usually fail? do they short or go open? Damn

[bklein]

Brian,
I think you should come to the condusion the unit is behaving as designed. You want different behavior for different purpose. So decide whether to sell it or modify it.
If it were me, I would first disable the TX and modify the RX to be symmetrical and rail to rail. The +/- can be reduced from V+/- but symmetrical. Re-enable the TX and see that things stay that way - if not, something there is unsymmetrically drawing power. Given that's good, adjust your sample timing shorter if you can. In fact, if the timing looks right as it is, I would think that is a good indication that the secondary pulse in the preamp output you are worried about is inherent to the design and understood (timing is skewed out alread to compensate). Compare the flyback timing with the Surf PI response and you can get an idea of whether the coil and/or its cable has too high capacitance. Heck, just substitute a quickly home-made or alternate coil. That all good, even the downstream circuits may not be good for your purpose - designed for slower coil movement(?). I guess you had some mention of this in an earlier thread.

[deemon]

I was having a look at an old pi machine I have that doesnt work very good,using my new scope I saw what looks like ringing on the preamp output.This machine has 330 ohm damping resistor so I didnt think it was ringing.I check the coil signal before the limiting diodes and it looks good,no ringing.I check at the peamp input and it looks good smooth tx pulse,flyback,and tailout to zero volts.But the output of the preamp shows tx pulse,flyback,cross zero opposite polarity,then tailout to zero.Let me post a picture for you.The yellow is the input,the red is the output,second pic yellow is coil signal before diodes,no ringing.

But what about the cables that connects the coils to the device ? Are the cables original or replaced ? If somebody replace the cable , and new cable does have bigger capacitance - it may cause underdamping of the coil , providing some "overshoot" of the pulse ...

[Brian Deese]

Post 13 shows the complete schematic. U2 SI7657? is a negative voltage converter by the looks of it. But can't find anything on that part so I must have the part number wrong. What do you have for V- here? I'm assuming it is the source for the comparator circuits.

I didn't really get an answer: Should the output if the 5534 be rail to rail? Post #15 shows alternating + and - preamp outputs that I imagine must be symmetrically sampled about the zero V reference point. Problem is that the amplitudes differ significantly. You've got a -1.5 division max. signal vs a +3 division one. I would think they should be the same peak amplitude if they are being sampled equivalently about the zero reference point.

Looks like I messed up about dynamic damping - instead these are the TX circuits.
So I go back to whether or not the positive and negative amplitudes out of the preamp should be identical - if yes, are these parts ok?

Lastly, as for the power draw - it will be the TX that draws the most. Is it enough to skew the negative rail? Whether or not the IC's are "military" or not is irrelevant.
In fact, these days what the heck does "military" mean? Made in China or not made in China?

Ok well have another look at post 37 and you can see the output of the 7661 is -4.7 volts,it is a voltage inverter yes.Then post 38 shows the math I have applied to the control circuitry of the regulators. 3.3 volts VC and -2.8volts VE. And that is why you see non symetrical preamp outputs yes,yes they are sampled near the 0v mark,one sample from the positive half cycle,and one from the negative half cycle are fed to the integrator.Look at the regs in the schematic and you will see the voltages are the correct levels,look into the mind of the designer to find out why they are different levels.I dont know.I am going to socket some of the key resistors and do some experimenting,I have to hit the ebay and do some shopping.I did dig up a 5 gallon bucket full of iron with this thing yesterday,when I get that additional transition out of the preamp output this thing will be an animal.

[Davor]

Whether or not the IC's are "military" or not is irrelevant.
In fact, these days what the heck does "military" mean? Made in China or not made in China?

Actually, military means extended temperature range, and it is required in case you are selling your equipment to, say, NATO. Otherwise, LM2902 is very same as very well known LM324, and that was the point - it draws very little current.

Just by looking at the components choice, and bipolar pulsing, you could say that primary role for this rig is demining.

[Mechanic]

Hi Barry,

Here's the datasheet : http://pdf1.alldatasheet.com/datashe...62-SI7661.html

It is a charge pump and could well be(or surrounding components) a source of trouble as to the low -voltage. The datasheet says voltage ranges from 4.5 to 20v. It would be worth checking the electrolytic caps around this area.

As to the 5534, yes it should go to the rails - a little bit, if the input signal and gain is high enough.

Cheers Mick

[bklein]

Post 13 shows the complete schematic. U2 SI7657? is a negative voltage converter by the looks of it. But can't find anything on that part so I must have the part number wrong. What do you have for V- here? I'm assuming it is the source for the comparator circuits.

I didn't really get an answer: Should the output if the 5534 be rail to rail? Post #15 shows alternating + and - preamp outputs that I imagine must be symmetrically sampled about the zero V reference point. Problem is that the amplitudes differ significantly. You've got a -1.5 division max. signal vs a +3 division one. I would think they should be the same peak amplitude if they are being sampled equivalently about the zero reference point.

Looks like I messed up about dynamic damping - instead these are the TX circuits.
So I go back to whether or not the positive and negative amplitudes out of the preamp should be identical - if yes, are these parts ok?

Lastly, as for the power draw - it will be the TX that draws the most. Is it enough to skew the negative rail? Whether or not the IC's are "military" or not is irrelevant.
In fact, these days what the heck does "military" mean? Made in China or not made in China?

[Brian Deese]

I wonder about the output not being symmetrical. Is the 5534 supposed to output to the negative rail? I can't determine this from my limited googling. What is the negative voltage IC as I can't make that out. I would think ideal would be symmetrical +\- supplies and rail to rail output here. I also wonder about your battery - a chance it needs to be more beefy than a 9v or AA cells. Try a bench supply. How old is this? Maybe weak ESR electrolytics.

Read post 38 there is no voltage reg ic. It has linear regs that are built with discreet components,I have already hashed out the voltage supply levels. Also notice this machine pulses the coil with VC and not B+. The positive reg transistor is the bigest part on the whole board.

[Davor]

... I also wonder about your battery - a chance it needs to be more beefy than a 9v or AA cells.

Actually not. Aside from 5534 all other opamps are military grade 324s, 393s, cmos and such, and I don't expect it to use more than 50mA at average, mostly for supplying the coil.

As I said - I like this design.

[bklein]

I wonder about the output not being symmetrical. Is the 5534 supposed to output to the negative rail? I can't determine this from my limited googling. What is the negative voltage IC as I can't make that out. I would think ideal would be symmetrical +\- supplies and rail to rail output here. I also wonder about your battery - a chance it needs to be more beefy than a 9v or AA cells. Try a bench supply. How old is this? Maybe weak ESR electrolytics.

[Davor]

First off, a log amp is my brain child, not a real implementation yet. However, your detector is capable of employing such idea - as is. It depends upon an inverting opamp frontend.

Second, this double coil business is a nice example of using exactly the same circuitry for opposite pulses. You must have both damping resistors and they must be the same value. In effect it is a 1:1 transformer, and with good coupling factor as well. I'd expect coupling better than 0.95.

From your plots I don't see any ringing, and you should try finding a root cause of your detector's bad behaviour elsewhere.

There are several trimmers to check. Did you confirm DC offset to be zero at 5534 output? Did you check timings?

[Brian Deese]

This is EXACTLY my log amp solution! Antiparallel diodes in feedback for a bipolar signal.

Brian, you are probing the inverting input of an opamp. If everything works in normal active regime it should read zero (0) volts against the non-inverting input, usually signal ground. In this case you don't have it. Opamp is not able to cope with input swing of a PI spike, so anti-parallel diodes kick in, and opamp saturates. Once current through diodes reaches equilibrium with current passing through feedback resistor, the opamp turn backs to normal working regime, but it takes some time, and at the very end of this transition it goes below zero - thanks to opamp's compensation circuitry and delays. Then it reaches normal active regime and voltage thereafter is zero. It is perfectly normal.

Now I have a question for you,in your bipolar pulsing detector do you use two coils wound in opposite directions or the same coil pulsed with opposite polarity? The thought that I am having is that if you have two coils encapsulated in the search head,in that close proximity,and they share a common ground does that add difficulty in critical damping? In this design it has two damping resistors of course, and the coil who is not active is held to zero with 390R. Just a thought,maybe not a good one.

[Brian Deese]

Hi Brian,

Why not try the diodes in the feedback loop. Just some general purpose diodes will do. Bav99 will do too. This should speed up the opamp recovery and enable an earlier sample, thus more sensitivity.

Cheers Mick

I will try it this afternoon,and by they way about the sample.I just wanted to let you guys know,In a previous post I said I would check the "secondary sample also" well there is no secondary sample.This machine feeds one sample from positive pulse and another from the consecutive negative pulse to the differential amp.Thats a change of pace right. And the sample can not start as soon as the other PI machines I have,you cant move the sample up to the egde of the saturation plateau.It can just include a few uS of the linear range of the amp and then the rest of the sample is at zero.I works similarly when you are inching it closer,when you go past a certain point it will not auto tune any longer and no output audio.I set the machine up yesterday for the best performance and its running at 4khz,with a 100uS TX width,and it likes a 40uS sample width the best. And another interesting thing about it since I have it running stable"albeit the preamp output" is when you place the coil on the ground and press the auto tune it oscillates at about 1hz,when you lift the coil it loses some ground influence and it goes quiet,sort of a silent search.Pretty neat I think.