Self sensing probes for oscilloscope

[G7PSK]

I have an Iwatsu SS-7804A oscilloscope that has auto sense for 1,10 and 100X probes. I do not have the original probes for the scope but have found that connecting the bnc to the ring at the base of the bnc on the scope with a 220k resistor gives 10X sense and a dead short gives a 100X while leaving no connection is 1X. What I would like to do is find the type of BNC that has the extra connection on it so that I can rig my probes to auto sense. Any one know what these type of BNC are called and where it is possible to source a few of them.

[edavid]

They are called readout encoding BNCs, or readout pin BNCs.  They are hard to find on their own.  You may have more luck buying some lower end used Tek probes such as P6062, P6102, P6105s, P6109.

[nctnico]

IIRC Testec has probes with sense pins but in my experience 1:1 / 1:10 switchable probes are much more convenient.

[edavid]

IIRC Testec has probes with sense pins but in my experience 1:1 / 1:10 switchable probes are much more convenient.

Tek P6062, P6063, P6127 and P6129 are switchable and have readout pins.  Maybe others too.

[G7PSK]

The tek probes that I have seen appear to have a sense pin that goes into a hole next to the bnc, the bnc's on the Iwatsu have an annular ring at the base which is no larger than the outside of the bnc that is insulated from the bnc and the case that makes the connection via a skirt or pin whatever it is the gap is only a couple of mm. I will post a picture of the scope connections.. I know that it is very easy to just do the maths with the probe settings, but having found that the scope can alter its read out automatically it would be rather nice to make it do so without having to spend a fortune on special probes. What I had in mind was a die cast box with a male and female bnc containing a suitable switch and resistor network so that I can just dial between 1,10 &100 times perhaps even via a tac button on the probe 

[edavid]

The tek probes that I have seen appear to have a sense pin that goes into a hole next to the bnc, the bnc's on the Iwatsu have an annular ring at the base which is no larger than the outside of the bnc that is insulated from the bnc and the case that makes the connection via a skirt or pin whatever it is the gap is only a couple of mm. I will post a picture of the scope connections.. I know that it is very easy to just do the maths with the probe settings, but having found that the scope can alter its read out automatically it would be rather nice to make it do so without having to spend a fortune on special probes. What I had in mind was a die cast box with a male and female bnc containing a suitable switch and resistor network so that I can just dial between 1,10 &100 times perhaps even via a tac button on the probe

You are mistaken about the Tek probes - there is no hole, they are the same as Iwatsu.  (Tek invented the system, and a few other makers copied it.)

The box is not really a good idea, since you would need to maintain the coaxial connection of the signal lead.  It would be better to add a switch or separate jack that connects to the sense lead inside the scope.

[alm]

Even adding a coaxial extension will increase the input capacitance and may make it impossible to compensate the probe.

[G7PSK]

Ok thanks for the info I will drop the idea for a multi switchable probe and find some tek probes.

[G7PSK]

I have just ordered some pogo pins, I will fix them to the out side of the BNC on the 100X probes I have as they are not switched and as the sense appears to be just short to ground or at least anything under 1K it should do the trick. I will post some pictures when the pins arrive most likely some time late in August as they are coming from China.

[alm]

Not sure if pogo pins are the ideal solution. The ring of standard BNC connector rotates if you attach it. Pogo pins are designed to dig into the PCB to improve the contact resistance. This might quickly wear out the surface. Tek uses just spring-loaded flat pins that result in much less wear. It's not like a few Ohms of contact resistance is going to affect their function.

[Andy Watson]

  on the 100X probes

If it's the 100X option, i.e. the zero-ohm short, that you want, I used a couple of turns of springy wire (not copper) around the shaft of the scope's bnc. Worked a treat for the Tek 475 10X option.

[G7PSK]

I found some pogo pins with round ends, thats why they have to come from China. I did think about bronze wire springs, i just thought the pins would look neater and more pucker. I would still like to make a probe switchable, but am still not sure how I will fit the switching especially the switch on the probe head. Most probes seem to have a reed switch in them so if I can fit another reed alongside that might be the way to go. I have some cheap probes coming in order to take them apart, no point in ruining good ones until I have the methodology sorted. 

[nctnico]

Ok thanks for the info I will drop the idea for a multi switchable probe and find some tek probes.

Actually the box isn't a bad idea. Maybe you could modify the scope to make the input range selectable. There should be room for a small on-off-on (3 position) toggle switch.

[G7PSK]

Ok thanks for the info I will drop the idea for a multi switchable probe and find some tek probes.

Actually the box isn't a bad idea. Maybe you could modify the scope to make the input range selectable. There should be room for a small on-off-on (3 position) toggle switch.

I was going to do that but it was pointed out that the extra capacitance might be too much for the scope. I cant see that the capacitance would be more than 1 o 2 pf if the box is kept small. What I was going to use was a box that is 25x25x185mm that way the one box will bridge the 2 inputs and the bnc socket and plug will practically touch each other across the box. I might still make the box to try it but for now I will adapt the 100x probes. What would be nice would to fit a network inside the box with remote switching, but that is something that requires more research, I am sure it could be done for 1 and 10 X but the voltages might be a bit high for the 100X as the circuit will have to be kept very small.

[nctnico]

Ok thanks for the info I will drop the idea for a multi switchable probe and find some tek probes.

Actually the box isn't a bad idea. Maybe you could modify the scope to make the input range selectable. There should be room for a small on-off-on (3 position) toggle switch.

I was going to do that but it was pointed out that the extra capacitance might be too much for the scope. I cant see that the capacitance would be more than 1 o 2 pf if the box is kept small. What I was going to use was a box that is 25x25x185mm that way the one box will bridge the 2 inputs and the bnc socket and plug will practically touch each other across the box. I might still make the box to try it but for now I will adapt the 100x probes. What would be nice would to fit a network inside the box with remote switching, but that is something that requires more research, I am sure it could be done for 1 and 10 X but the voltages might be a bit high for the 100X as the circuit will have to be kept very small.

Interrupting the probe cable is a bad idea indeed. The cable and the end connector are part of the probe's design! However if you can somehow make an external connecting to the sense ring then you can interface that to an external box.

[G7PSK]

I wasn't thinking of cutting the probe lead, the box would have a socket one side and plug the other, all the box would do is carry the sense circuit connections.

[nctnico]

Then you still insert a strange piece of cable and connectors in the probe's signal path. For low frequencies that is not a problem but when pushing to several tens of MHz the extra cable may impair the signal. Maybe you could make an extension plug which is very short.

[robrenz]

Mount a 3 position micro size switch in the scope next to each bnc. Wire internally for the 2 different resistances on 10x and 100x.  This way you are not interfering with the signal path at all. And when in 1x mode it will still accept the original probes with the auto sensing.