Pages: [1]   Go Down
Author Topic: RG-58 Relay Controlled by Arduino for Oscilloscope Inputs  (Read 911 times)
0 Members and 1 Guest are viewing this topic.
Offline Offline
Newbie
*
Karma: 0
Posts: 24
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Hi all, I'd like to control the inputs to an Oscilloscope by relaying RG-58 coax cables connected to two sets of transducers pairs (I'm doing TOF ultrasound calculations on tiny porous mediums).

The square wave pulser does 400V pulses so im worried that might be a problem. The pulse duration is only like 1/255 though. Anyway.

Is there such a thing as a coax relay controlled by a digital 5V input? And what might be the drawbacks of using one? What are some possible scenarios where I break the dirty expensive Oscilloscope/Square Wave Pulser?
Logged

Miami/Florida
Offline Offline
Sr. Member
****
Karma: 14
Posts: 342
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

You have given very little info on the set up you are attempting; but anyway I think you should avoid connecting the 400V pulses directly to the scope input. You can set simple attenuators using voltage dividers (let's say by 1/100) to work with a safe voltage at the scope input and if required, you can switch those signals safer using low voltage devices. The scope will work nicely with 4V at the input. Maybe you should use opto-couplers to send the commands from Arduino to the switching devices if you choose solid state ones. In that way, Arduino will be safe in case of HV failure. I think, the switching freq you are using is too high for conventional coil relays.
Logged

Perseverance is 90% of the solution. The remaining 10% is more perseverance.

Offline Offline
Newbie
*
Karma: 0
Posts: 24
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Thanks! I have an external Oscilloscope and square wave pulser.. I'm wondering what would be the best way to use the Arduino to digitally switch between two pairs of Coax cable.

Like.. I can't find any reasonably priced contraption that can mechanically switch between two different coax lines.
Logged

Offline Offline
Newbie
*
Karma: 0
Posts: 24
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

And just to clarify.. the switching frequency doesn't have to be super fast. When I was talking about the pulse duration I meant of the square wave pulser to indicate that although its 400V, the pulses are very-very short.
Logged

Miami/Florida
Offline Offline
Sr. Member
****
Karma: 14
Posts: 342
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

-What freq do you want to use for switching? That will determine which device you can use.
Logged

Perseverance is 90% of the solution. The remaining 10% is more perseverance.

Offline Offline
Newbie
*
Karma: 0
Posts: 24
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Well, as fast as possible would be good but I'm game for ideas that could take up to 1s.
Logged

Miami/Florida
Offline Offline
Sr. Member
****
Karma: 14
Posts: 342
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Sorry, you need to provide more info in order to help you.

-What scope are you using?.
-1 second is the switching signal period you are willing to have or its ON state? In other words, what's the frequency and duty cycle you want for your switching signal?
-When you mentioned 2 pairs, I guess is 4 coax cables and you want to switch 2 at a time to the scope 2 Channels inputs. Is that correct?
-If that's the case, are all 4 wires carrying 400V pulses? or is 2 TXs and 2 RXs?
-What's the pulse repetition freq (PRF) of the signal?.
-If you make a drawing of the set up you want that would help.

-If you use relays, the bouncing on the contacts will introduce noise. If the switching is done too fast, as your ON/OFF time gets closer to the duration of the bouncing the signal to noise ratio will be greatly deteriorated.
-If you switch 400V directly with relays there will probably be sparking in the contacts and you may need special HV relays to extend their life. Besides that you are risking damaging the scope input with such a HV.

Logged

Perseverance is 90% of the solution. The remaining 10% is more perseverance.

Offline Offline
Newbie
*
Karma: 0
Posts: 24
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

I made a video showing my setup and posted it to YouTube here:
http://www.youtube.com/watch?v=VEgH6jUwsFc&feature=youtu.be

-What scope are you using?.
A little bit more about the set up: A TDS3052C is connected to see the square wave pulse (TX) and the transducer response (RX). Both channels on the Oscilloscope are used but I don't use the square wave pulse signal for calculations. The front of the square wave pulser is connected to a set of ultrasound transducers. This project involves using Shear Wave transducers and Longitudinal Wave transducers. Set up as is, I have to disconnect the two transducers and reconnect the two transducers to get readings from both types of transducers to calculate my end goal: porosity.

-1 second is the switching signal period you are willing to have or its ON state? In other words, what's the frequency and duty cycle you want for your switching signal?
My goal is to come up with some sort of digitally controlled switch for the low-attenuation coax wires. I want to be able to set a digital pin to HIGH and have the relay switch transducers. I said 1 second before but that was because I wanted to give a high number - that's how long I figured the circuit would need to stabilize after the switch. I'm not sure what the duty cycle/frequency should be, I was thinking no frequency/pwm just manually controlled via the Arduino w/ Javascript button/script (I'm using BreakoutJS (Firmata based Arduino control via HTTP)) and interfaced with the Tektronix over Ethernet so all that stuff is taken care of on the computer.

The square wave pulser is set to always be on.

-When you mentioned 2 pairs, I guess is 4 coax cables and you want to switch 2 at a time to the scope 2 Channels inputs. Is that correct?
-If that's the case, are all 4 wires carrying 400V pulses? or is 2 TXs and 2 RXs?


Right now, (I think) only one transducer/scope pulses with the 400V and the other one receives it and there is a significant amount of attenuation. I'm actually not sure if I'm qualified to answer this because I know the square wave pulser is set to 400V but the Oscilloscope only reads 5V (with the Gain and Voltage at maximum). Maybe whoever bought the set up planned it out like this? IDK but square wave pulse channel on the Oscilloscope does say Probe Comp = 5V *Square Wave Sign* on the input.

-What's the pulse repetition freq (PRF) of the signal?.
The PRF settings read EXT, 100, 200, 500, 1K, 2K, and 5K. I'm not really sure what this is used for but I was thinking this setting might be useful to find the mass of these pills.

-If you use relays, the bouncing on the contacts will introduce noise. If the switching is done too fast, as your ON/OFF time gets closer to the duration of the bouncing the signal to noise ratio will be greatly deteriorated.
Not a huge deal - hopefully this project works out and I'm given a bigger budget to use on v2. Again, it's not super important that I switch the sets fast, even up to 10 seconds switch time would be acceptable although the faster the better and I'm trying to do this right the first time around.

-If you switch 400V directly with relays there will probably be sparking in the contacts and you may need special HV relays to extend their life. Besides that you are risking damaging the scope input with such a HV.
Here's what I was thinking might be one way to go that -might- address this issue. Assuming the receiving transducers only receive a small amount of voltage, maybe I can just have one relay that controls the receiving transducers and then just split the transmitting line so that both of the transmitting transducers are always running.

-If you switch 400V directly with relays there will probably be sparking in the contacts and you may need special HV relays to extend their life. Besides that you are risking damaging the scope input with such a HV.
It'd be nice but I'm also trying to get this done cheaply (under $100)... I only need it to last.. let's say 2000 switches -- enough time to test it, show it to my advisor, and the possible grant provider.

It may be important to note that Oscilloscope and square wave pulser have Ext Trig and Ext Trig In - is this where I should be connecting a ground from the Oscilloscope to the square wave pulser? Also, here's a screenshot of the Oscilloscope response when the Square wave pulser begins a pulse showing that the Oscilloscope is reading both a maximum of around 5V for both signals when the gain is at max and the voltage is at max.



Thanks for the help.
Logged

Miami/Florida
Offline Offline
Sr. Member
****
Karma: 14
Posts: 342
View Profile
 Bigger Bigger  Smaller Smaller  Reset Reset

Hi:

Thanks for the info. I begin to get the whole idea better now.

-Both transducers can be running at the same time; but if so, they should not be placed at the same time on the sample otherwise there maybe some cross talking and you will be picking up both signals at the receivers.

-Circuits grounds should be connected together. That is, your scope and pulse generator grounds should be common. External Inputs are not for that, they are used to trigger the units, do not connect them. In other words, the BNC connectors grounds on both units should be connected, so the circuit has a common ground.  If they are not connected now, that maybe the reason why you are reading only 5V instead of the 400v as the current path is not the units' ground and there is a huge voltage loss in the link. CAUTION: If you find now that their grounds are not  connected together, do not connect them as this may cause the voltage to go up to 400V at the scope input and damage it.

-I like your idea. Drive both TXs together by splitting the pulse generator signal and switch the RXs signals which are low voltage. CAUTION: when you split the signal the power you are suppliying to the TXs may decrease as you are increasing the load to the pulse generator and causing impedance mismatch. You need to try and see if the difference is acceptable or not.

-To switch the RXs signals you can built a metal box where you place 3 BNC connectors. 2 on one side and another in the opposite side. Inside the box you place your relay STDP (center contact switches to either one of the to ends). You connect your RXs signals to the 2 BNCs on one side and the scope input to the other BNC in the other side. The relay when activated switches the signal from either Rxs to the scope input. Place another low voltage jack to supply power and commutation signal to the relay. Seal the box. At 3MHz there should not be a problem with interference. You need to test and see the results. The good thing is that you are already familiar with what happens without this rig so you will be able to notice if there is any difference. Activate the relay from Arduino at 2 Hz square wave and check the results. You can easily change the pulses freq and duty cycle in the Arduino code to get the best results. You will get some noise when the relay switches; but that will quickly disappear and you will still have time enough to see the useful signal you want. It all depends on how fast the switching is done. The relay contacts once they close they should not affect signal transmission at 3MHz. Again trial and error until you get what you want.

-Built another metal box and set an attenuator using 2 simple resistors dividing the pulse generator voltage by 10. Feed the attenuated signal to the scope other input so you can have the Tx reference signal in the screen at all times to make the comparison with the received ones. The scope should work fine with 40V at the input.

Post here if you need more help. Also good if you post how it progresses
.
Good luck.
Logged

Perseverance is 90% of the solution. The remaining 10% is more perseverance.

Pages: [1]   Go Up
Jump to: