Analyzing and Understanding Waveforms on a TENS Unit

[JJ_023]

Need guidance in analyzing and understanding waveforms on a Transcutaneous Electrical Nerve Stimulation machine (TENS Unit).

I am asking the community to chime in and expand my understanding as well as learn a few things along the way.

I was playing with a TENS unit.  The unit is a medically prescribed unit.  I was curious about the parameters and what the actual unit put out.  Here are my findings. 

Maxed out the unit produces around 20V AC and 9.4 µA of current.  I used a fluke 87 V for this.

When I tried reading the Hz the meter was not capable.  This made me want to throw the unit on a scope and see the waveform.  This completely blew me away.  300+ volts peak to peak.  See the attached two images of the waveform I captured.

This particular unit has two adjustments.  One is called “PULSE RATE” which is actually the Hz.  It tops out at around 150 Hz and was accurately calibrated to the device.
The other control is called “Pulse Width”, which can be thought of as the duty cycle, but is labeled in a very confusing manner.

Why was the Fluke unable to read the Hz?  My guess is that the duty cycle was too low.

Could this waveform be re-created utilizing a signal generator from some of the Rigol or Siglent lower tier units that are around $600?  If not what would it take to re-create this type of signal?

Can this really put out 300 V peak to peak or am I misinterpreting something?

As usual thank you to anybody that comments or participates in this discussion.

[Kim Christensen]

Just need to pulse a coil/transformer to get some high voltage spikes. Simple 555 timer IC, or other pulse generator, driving a transistor which drives the primary of the transformer. The only difficult part is making the intensity the right level so it's not harmful or painful. Experiment on yourself first.
Essentially, it's a very low power TASER.. 

[ejeffrey]

I don't know anything about TENS or this circuit but 300 v pk-pk as measured by a scope probe sounds reasonable.  It takes a fair bit of voltage to get through the skin even for a small current.  However the available current / energy will be limited to a (hopefully) safe level.  In actual use the voltage will probably be lower due to the loading by the user.

You can't just program an AWG to make the waveform you see, that would be unsafe.

[JJ_023]

Just need to pulse a coil/transformer to get some high voltage spikes. Simple 555 timer IC, or other pulse generator, driving a transistor which drives the primary of the transformer. The only difficult part is making the intensity the right level so it's not harmful or painful. Experiment on yourself first.
Essentially, it's a very low power TASER.. 

Thank you for your input.   From my experience 555 circuits don't produce this type of waveform, so I am not sure how to do that.   There is also a negative aspect to this waveform that is not uniform.

[JJ_023]

I don't know anything about TENS or this circuit but 300 v pk-pk as measured by a scope probe sounds reasonable.  It takes a fair bit of voltage to get through the skin even for a small current.  However the available current / energy will be limited to a (hopefully) safe level.  In actual use the voltage will probably be lower due to the loading by the user.

You can't just program an AWG to make the waveform you see, that would be unsafe.

TENS units can be used for a lot of things such as nerve and muscle stimulation.  My question is why can't you program an AWG?   What would be unsafe about it?  The AWG is a much more Robust and precise piece of equipment then a TENS unit.

[Kim Christensen]

Just need to pulse a coil/transformer to get some high voltage spikes. Simple 555 timer IC, or other pulse generator, driving a transistor which drives the primary of the transformer. The only difficult part is making the intensity the right level so it's not harmful or painful. Experiment on yourself first.
Essentially, it's a very low power TASER.. 

Thank you for your input.   From my experience 555 circuits don't produce this type of waveform, so I am not sure how to do that.   There is also a negative aspect to this waveform that is not uniform.

Just google DIY TENs and you'll find some examples. This site has some simple designs and is less dodgy than some of the others I saw. A transformer or coil is what gives the high voltage spikes. The circuits could use some improvements, but look like they'd work as is.

[DTJ]

My question is why can't you program an AWG?   What would be unsafe about it?  The AWG is a much more Robust and precise piece of equipment then a TENS unit.

Just a guess, the output impedance of the AWG is probably a LOT lower than the TENs unit meaning the energy transferred into your body could be quite harmful.

[CaptDon]

I don't think there is a medically defined waveshape. I have worked on several different units for a local chiropractor and all they really were was charging an inductor and releasing the kick through a 50K pot. Most were stupid simple especially the battery operated units.

[JJ_023]

Just need to pulse a coil/transformer to get some high voltage spikes. Simple 555 timer IC, or other pulse generator, driving a transistor which drives the primary of the transformer. The only difficult part is making the intensity the right level so it's not harmful or painful. Experiment on yourself first.
Essentially, it's a very low power TASER.. 

Thank you for your input.   From my experience 555 circuits don't produce this type of waveform, so I am not sure how to do that.   There is also a negative aspect to this waveform that is not uniform.

Just google DIY TENs and you'll find some examples. This site has some simple designs and is less dodgy than some of the others I saw. A transformer or coil is what gives the high voltage spikes. The circuits could use some improvements, but look like they'd work as is.

I am not really looking to DIY a TENS unit.   I have access to several units already.   Oftentimes I just get curious about things so I ask questions and those answers usually lead to more questions.  I just enjoy the process of learning.

Thank you for sharing that link.  It's fairly interesting unfortunately it doesn't show the wave forms that are produced, but having played around with enough 555  timers it seems like it would be more of a square wave product then the wave I noticed.   Personally I don't know which one is better. 

[JJ_023]

My question is why can't you program an AWG?   What would be unsafe about it?  The AWG is a much more Robust and precise piece of equipment then a TENS unit.

Just a guess, the output impedance of the AWG is probably a LOT lower than the TENs unit meaning the energy transferred into your body could be quite harmful.

I think that is a pretty good guess and makes a lot of sense.  What would be a way to test the impedance of the TENS unit  so that it is relevant to this discussion?

[JJ_023]

I don't think there is a medically defined waveshape. I have worked on several different units for a local chiropractor and all they really were was charging an inductor and releasing the kick through a 50K pot. Most were stupid simple especially the battery operated units.

I think that you are correct and there is no medically defined wave shape.  Some of the units are incredibly simple in terms of their design.  It just never occurred to me that they would generate 300 volts AC peak to peak.

[TimFox]

I assume your 'scope waveforms were obtained using a 10 megohm (x10) probe.
If you take your DMM in ohms mode and measure between, say, your left hand fingers and right hand fingers, you will probably see a much lower resistance (maybe 10 to 100 k\$\Omega\$), depending on your skin moisture, which will load down the TENS unit output substantially.

[thm_w]

As mentioned, its 300V with no load, once you load it down you'll see there is not much total power behind it. Resistance varies so much: dampness of the skin, area of the skin, condition and size of the electrode pads, so its good to have variable drive strength.

There isn't a defined waveshape, but there are arguably more optimal waveforms and frequencies, for whatever you are trying to accomplish.
Low frequencies produce more muscle movements, high frequency less noticeable, more tingling sensation. So for pain relief it makes sense to use a higher frequency, as it can be on for a long time without being overly distracting/tiring.

https://link.springer.com/article/10.1007/BF02371250
https://karger.com/sfn/article-abstract/44/5-6/330/333737/Transdermal-Transcutaneous-Electric-Nerve pain relief
https://link.springer.com/article/10.1007/s00421-002-0711-4 muscle stimulation
https://sci-hub.se/https://www.tandfonline.com/doi/full/10.1586/ern.11.48

In the last article there is a medical piezo barbeque lighter listed, that will generate up to 15kV unloaded.

[JJ_023]

I assume your 'scope waveforms were obtained using a 10 megohm (x10) probe.
If you take your DMM in ohms mode and measure between, say, your left hand fingers and right hand fingers, you will probably see a much lower resistance (maybe 10 to 100 k\$\Omega\$), depending on your skin moisture, which will load down the TENS unit output substantially.

You are correct about the assumptions on the probe.  When you say " see a much lower resistance" compared to what?   I'm not following your train of thought.   Sorry.

[JJ_023]

As mentioned, its 300V with no load, once you load it down you'll see there is not much total power behind it. Resistance varies so much: dampness of the skin, area of the skin, condition and size of the electrode pads, so its good to have variable drive strength.

There isn't a defined waveshape, but there are arguably more optimal waveforms and frequencies, for whatever you are trying to accomplish.
Low frequencies produce more muscle movements, high frequency less noticeable, more tingling sensation. So for pain relief it makes sense to use a higher frequency, as it can be on for a long time without being overly distracting/tiring.

https://link.springer.com/article/10.1007/BF02371250
https://karger.com/sfn/article-abstract/44/5-6/330/333737/Transdermal-Transcutaneous-Electric-Nerve pain relief
https://link.springer.com/article/10.1007/s00421-002-0711-4 muscle stimulation
https://sci-hub.se/https://www.tandfonline.com/doi/full/10.1586/ern.11.48

In the last article there is a medical piezo barbeque lighter listed, that will generate up to 15kV unloaded.

Thank you for the articles.  The first ones really don't say much they are abstracts.  Last one is quite interesting and goes into detail.

So none of the wave forms describe what I saw on the oscilloscope.   It would be really cool if they included images of the actual wave shapes.

So my question still stands.  Why couldn't you use a AWG for TENS?    This would probably give Researchers a lot of flexibility in exploring this particular topic.  This to me is a clear example of where their would be Synergies if different scientific bodies collaborated with one another in research.

[thm_w]

What Tim said is you'll see a much lower resistance on your skin, compared to your scope probe input impedance (10MOhm).
So you can test with 10k to 100k resistor instead of just the 10MOhm of the scope.

As already explained above, you can't use a AWG as it does not produce 300V. You'd need to amplify its output, and then make sure it is intrinsically safe by limiting the current, etc.
You can find waveforms in the first articles if you use scihub.

[T3sl4co1l]

An average off-the-shelf AWG alone isn't likely very useful, but ask yourself: if the voltage is high, and the impedance is high, and the load resistance is highly variable, what kind of source characteristic might be useful to address that?  Something that adjusts its voltage to suit the load, perhaps?  A constant current source, say?  But not a fixed one, dependent -- a transconductance amplifier (gain = output current / input voltage), perhaps?

If you think in terms of an AWG plus a suitable amplifier, the opportunities become much more apparent.

Tim

[JJ_023]

What Tim said is you'll see a much lower resistance on your skin, compared to your scope probe input impedance (10MOhm).
So you can test with 10k to 100k resistor instead of just the 10MOhm of the scope.

As already explained above, you can't use a AWG as it does not produce 300V. You'd need to amplify its output, and then make sure it is intrinsically safe by limiting the current, etc.
You can find waveforms in the first articles if you use scihub.

I will definitely try playing with different resistance values as you suggested.

Thank you for the scihub reference, I was unfamiliar. I primarily use pub med.

[JJ_023]

An average off-the-shelf AWG alone isn't likely very useful, but ask yourself: if the voltage is high, and the impedance is high, and the load resistance is highly variable, what kind of source characteristic might be useful to address that?  Something that adjusts its voltage to suit the load, perhaps?  A constant current source, say?  But not a fixed one, dependent -- a transconductance amplifier (gain = output current / input voltage), perhaps?

If you think in terms of an AWG plus a suitable amplifier, the opportunities become much more apparent.

Tim

Keep in mind I am fairly novice.  So the way I understand your suggestion is the AWG will be used to create the desired waveform and its parameters.   Then run through an amplifier that will give it the desired voltage output at a predefined constant current.   Am I getting this correctly?

[RJSV]

I think the recent advice is simply saying that your waveform, into a 10 mohm load (probe),
isn't as valid, as a 10k or 50 k ohm load (body).
That being 200 to 1000 times lower of a load.
   So, maybe try to see what the waveform shows, using a realistic load, 10k to 100k.   That likely will drastically lower that 300 volt reading.

[CaptDon]

Bottom line, can a store bought unit produce 300 volts peak into an open circuit. Answer, yes it can. Does it need to limit the available current to a safe level. Answer, yes it does. Most of the units I repaired had a small transformer / inductor that was used to produce the 'kick'. The size of the core and the charge circuit limited the available discharge energy. On the output of the inductor was either a 50K or 100K pot wrongly named 'volume'. The wiper of the pot was the adjustable output and I believe also had a 10K resistor in series to the output lead. All units had an adjustable pulse repetition rate and some could do alternating pulse polarities. Even the 'fancy looking' units didn't have much inside of them. They just looked impressive. When it comes right down to it, the unit is barely more than a very feeble fence charger with a 'volume' control.

[JJ_023]

I think the recent advice is simply saying that your waveform, into a 10 mohm load (probe),
isn't as valid, as a 10k or 50 k ohm load (body).
That being 200 to 1000 times lower of a load.
   So, maybe try to see what the waveform shows, using a realistic load, 10k to 100k.   That likely will drastically lower that 300 volt reading.

Thank you for your suggestion.  I was actually just playing with that idea.   The unit is designed for medical use.   So I used myself as the actual load.   Here's the interesting part, the voltage went down to about 200 V AC peak to peak but the current increased dramatically to about 60 µA.  The shape of the waveform remained unchanged to what I posted initially.

For what it's worth I measured Just under 2 Mega ohm between the two test points on my skin.

[JJ_023]

Bottom line, can a store bought unit produce 300 volts peak into an open circuit. Answer, yes it can. Does it need to limit the available current to a safe level. Answer, yes it does. Most of the units I repaired had a small transformer / inductor that was used to produce the 'kick'. The size of the core and the charge circuit limited the available discharge energy. On the output of the inductor was either a 50K or 100K pot wrongly named 'volume'. The wiper of the pot was the adjustable output and I believe also had a 10K resistor in series to the output lead. All units had an adjustable pulse repetition rate and some could do alternating pulse polarities. Even the 'fancy looking' units didn't have much inside of them. They just looked impressive. When it comes right down to it, the unit is barely more than a very feeble fence charger with a 'volume' control.

I think this is a pretty good summation. Thank you for that.   The electronics part might be similar to a fence charger but the actual devices I think are a lot more sophisticated in the design aspect than that.   I think there is a lot to be said for the shape of the waveform which it seems a lot of different medical TENS  units try to incorporate based on different theories of what those do.   For example nerve stimulation versus muscle stimulation versus depth of stimulation. 

[mzzj]

I don't know anything about TENS or this circuit but 300 v pk-pk as measured by a scope probe sounds reasonable.  It takes a fair bit of voltage to get through the skin even for a small current.  However the available current / energy will be limited to a (hopefully) safe level.  In actual use the voltage will probably be lower due to the loading by the user.

You can't just program an AWG to make the waveform you see, that would be unsafe.

Resistance gets suprisingly low with the proper tens pads. 500-1000ohms per limb is good guess with large 2*2” pads. I have measured less than 500 ohms in some cases.
IEC 60601-2-10 Standard calls for 500 ohm test load and staggering 500vpp open circuit voltage. Typical cheapo home unit delivers around 100vp or 200vpp.

Above mentioned standard has further rules on output current and pulse energy. Allowed limits are much higher than one might initially think.

[JJ_023]

I don't know anything about TENS or this circuit but 300 v pk-pk as measured by a scope probe sounds reasonable.  It takes a fair bit of voltage to get through the skin even for a small current.  However the available current / energy will be limited to a (hopefully) safe level.  In actual use the voltage will probably be lower due to the loading by the user.

You can't just program an AWG to make the waveform you see, that would be unsafe.

Resistance gets suprisingly low with the proper tens pads. 500-1000ohms per limb is good guess with large 2*2” pads. I have measured less than 500 ohms in some cases.
IEC 60601-2-10 Standard calls for 500 ohm test load and staggering 500vpp open circuit voltage. Typical cheapo home unit delivers around 100vp or 200vpp.

Above mentioned standard has further rules on output current and pulse energy. Allowed limits are much higher than one might initially think.

Thank you for all the data points. After reading your post I decided to check the resistance with the pads.   The pads measure 1.5 inches square.   I placed the pads on my skin within 6 inches of one another, similar to the way that they would be utilized.   I got a reading of 600 K ohms (plugging the multimeter probes into the back of the pads).  So my results are not lining up with your claims.  What is it that I am doing wrong in my measurement's?