Problem with TGSL EDU

[knutdi]

Hi compadre, i use a plastic case, i will take a metal case and build all in, but i think that this is not the problem, with my plastic case, when only touch or move the coil wire a beep sounds

[satdaveuk]

If you have used a plastic case make sure you have bonded the coil plug/socket, and if you have used metal housing pots bond them also, if you done all that everything should be ok.
If the pots you used have plastic housings just leave as they are.
If your using USB cable that is not a problem unless you have used the cheap thin crapy stuff which is no good to man nor beast.

[knutdi]

Hello satdaveuk, I used metal pots, and housings are connected to earth, do you mean by "make sure the coil bonded Have you plug / socket"? you mean to ground the connector that connects the coil with the plastic case?

I have used various types of USB cable and all the same thing happens, I have even connected the coil directly to the board.

Tomorrow i will tested all.

[Mullihaka]

If your using USB cable that is not a problem unless you have used the cheap thin crapy stuff which is no good to man nor beast.

I'm using that 3mm thick USB cable, made for Nokia phones. I have just cut one of the connectors away. What's in thin cables to make them avoidable?

They do add 1nF capacitance but that's not a problem as I tweak the other capacitors to achieve the correct frequencies.

Crosstalk?

[simonbaker]

I'm using that 3mm thick USB cable, made for Nokia phones. I have just cut one of the connectors away. What's in thin cables to make them avoidable?

They do add 1nF capacitance but that's not a problem as I tweak the other capacitors to achieve the correct frequencies.

Crosstalk?

I think they are probably OK. The shields may not be as good, but I would need to see actual proof where one cable works better. On the other hand, it never hurts to over-engineer if you don't mind paying the money. I like to see how cheap I can go.

One thing that may matter is the connector. I noticed that USB connectors seem to be very sensitive to touch. Maybe more expensive cables have more solid connector.

Another thing is that really cheap cables may have no shield at all, although I have never seen that with USB cable. I have seen that with other cheap cables like SD video cables I think.

On the other hand, I looked at a Tesoro Compadre cable, and I did not see any shield at all!

Crosstalk may not matter much either. TX signal crosstalk may shift null phase a little, but that may not matter at all if you null correctly.

So I will believe proven results.

-SB

[satdaveuk]

Hello satdaveuk, I used metal pots, and housings are connected to earth, do you mean by "make sure the coil bonded Have you plug / socket"? you mean to ground the connector that connects the coil with the plastic case?

I have used various types of USB cable and all the same thing happens, I have even connected the coil directly to the board.

Tomorrow i will tested all.

The actual plug casing needs to be grounded, although theres no other phisical connection to it, must be done ok.
But if your wiring directly to the pc then oviously thats not your problem.
Are you sure that your getting a good earth bonding between your USB screen and earth on your PCB.
Check continuety between your negative battery feed and earthing point where coil screen is connected, there should be no resistance at all, may sound simple but it will give you all the simptoms you put on here.

[satdaveuk]

I think they are probably OK. The shields may not be as good, but I would need to see actual proof where one cable works better. On the other hand, it never hurts to over-engineer if you don't mind paying the money. I like to see how cheap I can go.

One thing that may matter is the connector. I noticed that USB connectors seem to be very sensitive to touch. Maybe more expensive cables have more solid connector.

Another thing is that really cheap cables may have no shield at all, although I have never seen that with USB cable. I have seen that with other cheap cables like SD video cables I think.

On the other hand, I looked at a Tesoro Compadre cable, and I did not see any shield at all!

Crosstalk may not matter much either. TX signal crosstalk may shift null phase a little, but that may not matter at all if you null correctly.

So I will believe proven results.

-SB

HaHa
Simon I like your ways inocent till proved guilty yeh.
I haven't proved as in a certainty but what I did do was hopefully improve just a tad more stabilty.
The USB3 nasa cable conductors are thicker, theres better screening, and the over all cable is more robust.
Maybe it is over engineering but on the other hand one more thing to tick off the list and for the sake of a couple of quid more for a 5m lead that would easily wire 4 detectors plus a test lead is either here or there.
we seem to be getting to a point on where theres no real improvements coming up on here apart from Silverdogs sense mod - to supply and ivionics preset in the TX circuit which have gave the TGSL some extra wallop in the field.
unless there is one I dont know about, if there is someone please tell me.
My next step is to do away with all IC sockets because a circuit of this type will be better without them im pretty sure of that. but again needs to be proved and by doing that it will open up another debate on quality build of IC sockets, theres lots of crappy ones out there, and maybe thats where a few peoples problems have been.
Anyway food for thought and keep your work coming Simon

Regards

[Jerry]

My next step is to do away with all IC sockets because a circuit of this type will be better without them im pretty sure of that. but again needs to be proved and by doing that it will open up another debate on quality build of IC sockets, theres lots of crappy ones out there, and maybe thats where a few peoples problems have been.
Anyway food for thought and keep your work coming Simon
Regards

Hi:

I have heard the debate on IC sockets ever since IC's became popular. I can see the theory behind it but I have never been able to prove one way or the other that there is a difference. But then I am not using them for microwave circuits either

I use sockets just because it is convenient but see no reason not to solder them in once a circuit is proven and debugged.

Just my take on it.

Jerry

[Qiaozhi]

My next step is to do away with all IC sockets because a circuit of this type will be better without them im pretty sure of that. but again needs to be proved and by doing that it will open up another debate on quality build of IC sockets, theres lots of crappy ones out there, and maybe thats where a few peoples problems have been.
Anyway food for thought and keep your work coming Simon

Regards

In this type of circuit it is the comparators that may have problems with IC sockets. That would be the first thing to try. As the input signals are always hovering around the threshold point, this can cause instability, and there is no hysteresis in this design.

[simonbaker]

HaHa
Simon I like your ways inocent till proved guilty yeh.
I haven't proved as in a certainty but what I did do was hopefully improve just a tad more stabilty.
The USB3 nasa cable conductors are thicker, theres better screening, and the over all cable is more robust.
Maybe it is over engineering but on the other hand one more thing to tick off the list and for the sake of a couple of quid more for a 5m lead that would easily wire 4 detectors plus a test lead is either here or there.
we seem to be getting to a point on where theres no real improvements coming up on here apart from Silverdogs sense mod - to supply and ivionics preset in the TX circuit which have gave the TGSL some extra wallop in the field.
unless there is one I dont know about, if there is someone please tell me.
My next step is to do away with all IC sockets because a circuit of this type will be better without them im pretty sure of that. but again needs to be proved and by doing that it will open up another debate on quality build of IC sockets, theres lots of crappy ones out there, and maybe thats where a few peoples problems have been.
Anyway food for thought and keep your work coming Simon

Regards

All good advice there! As Ivconic said, "mechanically sound is electrically sound", or something like that. There is nothing worse than flaky connections or cheesy shielding.

On my long todo list is a good investigation of noise sources to try to see where it is worth spending the money. Of course any "myth-busting" that happens along the way is fun.

Cheers,

-SB

[knutdi]

The actual plug casing needs to be grounded, although theres no other phisical connection to it, must be done ok.
But if your wiring directly to the pc then oviously thats not your problem.
Are you sure that your getting a good earth bonding between your USB screen and earth on your PCB.
Check continuety between your negative battery feed and earthing point where coil screen is connected, there should be no resistance at all, may sound simple but it will give you all the simptoms you put on here.

Hi, today i have tested all ground points, and all are good.
I also tested the feed point, the only thing wrong is on pin 4 of the LM393, LF353, LM358 and LM308, where should have -5V, I have -6v.

Someone with EDU TGSL could measure to verify that voltage is in this pin???

Thank you

[homefire]

As per this schemic your six volts is good!

I thought you found the problem and had this one up and running?

[knutdi]

Hi homefire, in my schematic this point is -5V, i attachment this.

unfortunately I have not found the problem, I checked everything and the problem persists

[hugo]

hi knutdi,
I've got the TGSL-EDU and the negative voltage is -6V too.
There is nothing wrong about it.
If you really want to have -5V, just add a 79L05 regulator ... but it's useless.

About the touchy cable, I've got the same problem with the USB2 cable.
Try another kind of cable, fix it along the lower stem and check if the connector is well grounded.

[knutdi]

Hi Hugo, thank for you reply.

If -6V is good dont need use an 7905.

I use several usb cable, all is good grounded less conector, but i test without cable, connecting coil cable directly to TSL board, and the problem persist, when I approach my hand to TGSL board components or wiring of the POT, a beep sounds.

Look my videos on post 3.

Can you measuring resistance of TX and RX??
I have:

RX 25.8 ohms --- 107 laps

TX 23.8 ohms --- 98 laps