Jump to content

Galvanic Isolators

William Martin

By William Martin
in Boat Equipment

 Share

Featured Posts

chris w Veteran II

chris w

Posted
Posted

I set up my Sterling 40A Switched Mode Charger and inserted 4 diodes in GI configuration into the earth lead. I then measured and observed the resulting waveforms across the diode pack on a scope.

 

I saw 60mv pk-pk square wave (= 30mv rms for a square wave) at ~30KHz and 50% duty cycle (which is the switching frequency of the charger). I tried the charger off-load and on-load (ie: charging a 110AH battery) and the result was the same. The output DC also showed the identical ripple, viz: 60mv pk-pk which in an output of around 15v dc equates to an impressive -54dB.

 

From those measurements, my particular charger does not seem to pose a problem. It would be interesting to compare results with other brands/types etc.

 

Chris

Link to comment
Share on other sites
Yoda Experienced

Yoda

Posted
Posted

I set up my Sterling 40A Switched Mode Charger and inserted 4 diodes in GI configuration into the earth lead. I then measured and observed the resulting waveforms across the diode pack on a scope.

 

I saw 60mv pk-pk square wave (= 30mv rms for a square wave) at ~30KHz and 50% duty cycle (which is the switching frequency of the charger). I tried the charger off-load and on-load (ie: charging a 110AH battery) and the result was the same. The output DC also showed the identical ripple, viz: 60mv pk-pk which in an output of around 15v dc equates to an impressive -54dB.

 

From those measurements, my particular charger does not seem to pose a problem. It would be interesting to compare results with other brands/types etc.

 

Chris

 

Should the amount of non dc voltage output, be part of the specification stated on the lable of sm power supplies?

Link to comment
Share on other sites
Gibbo Veteran II

Gibbo

Posted
Posted

I set up my Sterling 40A Switched Mode Charger and inserted 4 diodes in GI configuration into the earth lead. I then measured and observed the resulting waveforms across the diode pack on a scope.

 

I saw 60mv pk-pk square wave (= 30mv rms for a square wave) at ~30KHz and 50% duty cycle (which is the switching frequency of the charger). I tried the charger off-load and on-load (ie: charging a 110AH battery) and the result was the same. The output DC also showed the identical ripple, viz: 60mv pk-pk which in an output of around 15v dc equates to an impressive -54dB.

 

From those measurements, my particular charger does not seem to pose a problem. It would be interesting to compare results with other brands/types etc.

 

Chris

 

Interesting.....

 

Did you not get any 50Hz on there? Did you try it with the diodes out of circuit (i.e. with the earth floating)?

 

It's *especially* interesting as that particular charger uses a boost converter active power factor corrector as the first stage so there's no way on this planet you should have just seen a 50% duty cycle square wave on there. At the very least it should have been frequency or mark/space modulated at 50Hz.

 

That charger also therefore has *two* switching freqnecies. How did you decide the switching frequency was 30Khz? That's *extremely* low for a charger of that power/weight ratio..

 

The mark space ratio also should have changed depending upon the load yet you state it didn't.

 

Also the first filter stage has 2 X 10nF Y caps then a differential mode choke then 2 more 10nF Y caps so there should have been a substantial 50Hz component on there. Certainly enough to swamp the high frequency and make it almost impossible to see on a scope.

 

Are you sure about this?

 

Gibbo

Link to comment
Share on other sites
Gibbo Veteran II

Gibbo

Posted
Posted

Should the amount of non dc voltage output, be part of the specification stated on the lable of sm power supplies?

 

On many SM chargers it's in the manual specified as either an rms ac voltage or as being "XdB down" or "-XdB". I think they phrase it that way so that no one except engineers can understand it :closedeyes:

 

As long as it isn't too high (which none of them are) it simply isn't an issue. Up to about 100mV is perfectly fine in all installations. 250mV fine in most.

 

Some of the older line frequency chargers had up to 2 volts ripple on a 12 volt charger and even that rarely caused a problem.

 

Gibbo

Link to comment
Share on other sites
chris w Veteran II

chris w

Posted
Posted (edited)

Did you not get any 50Hz on there? Did you try it with the diodes out of circuit (i.e. with the earth floating)? Yes, with the earth floating, I only saw mains.

 

It's *especially* interesting as that particular charger uses a boost converter active power factor corrector as the first stage so there's no way on this planet you should have just seen a 50% duty cycle square wave on there. At the very least it should have been frequency or mark/space modulated at 50Hz.Well it's as close to 50% as makes no difference and it was not variable. This morning, after being ON all night and the charger having reverted to Float/Power Pack mode, it was exactly the same.

 

That charger also therefore has *two* switching freqnecies. How did you decide the switching frequency was 30Khz? That's *extremely* low for a charger of that power/weight ratio.. The mark space ratio also should have changed depending upon the load yet you state it didn't. I have a good 100MHZ Tektronix scope with (inter alia) frequency readout AND I measured it on the graticule just to be certain it was indicating the right thing. The squarewave wasn't super clean and there were obviously some higher frequency harmonics/components on to which I couldn't trigger If the mark/space ratio did change, it obviously wasn't by any readily noticeable amount. I have seen this figure of 30KHz written down somewhere amongst the very copious info on the Sterling site; I'll have a dig around to try to find it.

 

 

Also the first filter stage has 2 X 10nF Y caps then a differential mode choke then 2 more 10nF Y caps so there should have been a substantial 50Hz component on there. Certainly enough to swamp the high frequency and make it almost impossible to see on a scope. Are you sure about this?

I can only tell you what I saw and there was no mains measurable unless I lifted the earth completely.

 

 

 

Chris

Edited by chris w
Link to comment
Share on other sites
Gibbo Veteran II

Gibbo

Posted
Posted

qué???

 

Post #180 in this thread is a mess. Red writing everywhere, bracketed quote marks everywhere. Font brackets and all sorts. When I click on reply it goes to a webpage that doesn't exist!

 

All the other posts on all the other threads are fine. Just #180

 

Gibbo

Link to comment
Share on other sites
Gibbo Veteran II

Gibbo

Posted
Posted

Well it looks OK to me. I did have a couple of quote and font errors initially but I fixed them straightaway.

 

I hope I don't get a detention for this :closedeyes:

 

LOL

 

It's a mess here. I can't reply to it.

 

How do you explain the 50Hz disappears when you connect the diodes? They are open circuit until they conduct. So apart from the diode leakage the earth is still disconnected below the clampting voltage of the GI. It makes no sense. At all. Can you play with some resistors in the earth lead and see what value you need to swamp the 50Hz. It seems the diode leakage (of what? a few 100 MOhms?) is swamping the 50Hz? Which seems very odd as it's coming from from 230 volts via 20nF of capacitance.

 

Gibbo

Link to comment
Share on other sites
blackrose Veteran II

blackrose

Posted
Posted

Post 180 looks ok to me. I've quoted it below to see if it's any better for you Gibbo.

 

Did you not get any 50Hz on there? Did you try it with the diodes out of circuit (i.e. with the earth floating)? Yes, with the earth floating, I only saw mains.

 

It's *especially* interesting as that particular charger uses a boost converter active power factor corrector as the first stage so there's no way on this planet you should have just seen a 50% duty cycle square wave on there. At the very least it should have been frequency or mark/space modulated at 50Hz.Well it's as close to 50% as makes no difference and it was not variable. This morning, after being ON all night and the charger having reverted to Float/Power Pack mode, it was exactly the same.

 

That charger also therefore has *two* switching freqnecies. How did you decide the switching frequency was 30Khz? That's *extremely* low for a charger of that power/weight ratio.. The mark space ratio also should have changed depending upon the load yet you state it didn't. I have a good 100MHZ Tektronix scope with (inter alia) frequency readout AND I measured it on the graticule just to be certain it was indicating the right thing. The squarewave wasn't super clean and there were obviously some higher frequency harmonics/components on to which I couldn't trigger If the mark/space ratio did change, it obviously wasn't by any readily noticeable amount. I have seen this figure of 30KHz written down somewhere amongst the very copious info on the Sterling site; I'll have a dig around to try to find it.

Also the first filter stage has 2 X 10nF Y caps then a differential mode choke then 2 more 10nF Y caps so there should have been a substantial 50Hz component on there. Certainly enough to swamp the high frequency and make it almost impossible to see on a scope. Are you sure about this?

I can only tell you what I saw and there was no mains measurable unless I lifted the earth completely.

Chris

Link to comment
Share on other sites
smileypete Veteran II

smileypete

Posted
Posted

Hi,

 

I think my earlier message has been misinterpreted, so I'll reiterate and clarify some of it:

 

The results I would like to see is some real world measurement including:

 

The type and size of boat.

The type or design of the GI.

The equipment used which generates RFI.

The amount of AC and most importantly DC conducted by the GI.

In the last sentance I 'd better have said the amount of AC current and DC current conducted by the GI.

 

It's not the voltage across the GI but any DC current flowing through it that's important. It's DC current flowing through the earth connection that contributes to galvanic corrosion.

 

Without some measurement of this, there's no way of knowing if amount of corrosion caused is negligable or significant.

 

To put this in perpective, an AC voltage of around 1v peak (0.7v rms) would allow, on average, about 60% of the galvanic action to take place that would have taken place had there been no GI in place.

 

This just doesn't look right to me, surely it depends on the source impedance of the AC voltage.

 

cheers,

Pete.

Link to comment
Share on other sites
chris w Veteran II

chris w

Posted
Posted

Pete

 

The point is with a GI is that in effect it is just a switch (albeit an electronic one).

 

If the applied voltage across the GI is less than 1.2v, the GI is effectively an open switch and nothing can flow through it. So galvanic currents, which represent a maximum voltage of 0.8v (or so we are told in the professional literature) are blocked. That's why GI's are specified at 1.2v as this equals the forward voltage drop of two silicon diodes in series.

 

If however an AC voltage is present on the earth line it will swing up and down, as AC does, and if any part of that waveform is above 1.2v then it will cause the GI diodes to conduct as if they were a closed switch. They will only conduct for that part of the AC cycle that is above 1.2v.

 

Now, suppose we have a 0.8v DC galvanic voltage present as well. In this case, once the AC exceeds 0.4v the diodes conduct as the total voltage on them is a total of 1.2v. For that part of the AC cycle above 0.4v in this instance, there is no galvanic protection.

 

Obviously if the AC was a couple of volts or more, there would effectively be no GI protection at all as the diodes would be conducting for virtually all of the AC cycle. The impedance of the diodes, once biased into conduction, is very small and thus the limiting factor on AC current would be the AC source impedance and not the diodes. However, even a minute current will bias the diodes ON once the forward voltage drop is exceeded.

 

 

Chris

Link to comment
Share on other sites
Gibbo Veteran II

Gibbo

Posted
Posted

Pete

 

The point is with a GI is that in effect it is just a switch (albeit an electronic one).

 

If the applied voltage across the GI is less than 1.2v, the GI is effectively an open switch and nothing can flow through it. So galvanic currents, which represent a maximum voltage of 0.8v (or so we are told in the professional literature) are blocked. That's why GI's are specified at 1.2v as this equals the forward voltage drop of two silicon diodes in series.

 

If however an AC voltage is present on the earth line it will swing up and down, as AC does, and if any part of that waveform is above 1.2v then it will cause the GI diodes to conduct as if they were a closed switch. They will only conduct for that part of the AC cycle that is above 1.2v.

 

Now, suppose we have a 0.8v DC galvanic voltage present as well. In this case, once the AC exceeds 0.4v the diodes conduct as the total voltage on them is a total of 1.2v. For that part of the AC cycle above 0.4v in this instance, there is no galvanic protection.

 

Obviously if the AC was a couple of volts or more, there would effectively be no GI protection at all as the diodes would be conducting for virtually all of the AC cycle. The impedance of the diodes, once biased into conduction, is very small and thus the limiting factor on AC current would be the AC source impedance and not the diodes. However, even a minute current will bias the diodes ON once the forward voltage drop is exceeded.

Chris

 

And for once I'll agree with Chris. See, anything is possible given time.

 

Gibbo

Link to comment
Share on other sites
Osprey Sprinkler Explorer

Osprey Sprinkler

Posted
Posted

Quick and easy questions:

 

What is a typical resistance between a narrowboat hull and a neighbouring narrowboat hull in a marina? No shore connections, just through the water.

 

What is a typical resistance between a narrowboat hull and true earth? Via the canal water, again no shore connections.

 

Just wanted to get an idea of magnitudes.

 

Ashley

Link to comment
Share on other sites
Gibbo Veteran II

Gibbo

Posted
Posted

Quick and easy questions:

 

What is a typical resistance between a narrowboat hull and a neighbouring narrowboat hull in a marina? No shore connections, just through the water.

 

What is a typical resistance between a narrowboat hull and true earth? Via the canal water, again no shore connections.

 

Just wanted to get an idea of magnitudes.

 

Ashley

 

If only it was that simple!

 

It depends what voltage you measure it with, believe it or not. Which kinda makes it difficult to answer.

 

Gibbo

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share
Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.