Isolation Transformers

[MidLandOne]

Don't bore me with your self claimed credentials - surprising as it may be some of us have a little education and experience too.

So you are saying that if the windings are in parallel with a TCB in series with each winding like you claimed they were, so like this -

|---------TCB-------winding----|
|---------TCB-------winding----|

then the windings are in parallel but the TCB's are not in parallel .

My claim was and remains that they are and I suspect most would agree with me. Not much more that I can say as it is difficult to add much more to the obvious so I will leave it to you to bamboozle us with your credentials instead .

[s/v Jedi]

Midlandone,

Sorry to have offended you. You must understand that I try to help some people with their questions on Isolation transformers. I did put quit some effort in it too. If you think that I am wrong, or mis-informing the community, than it is to you to correct that with the right information as with links on the web, or leave me in peace. I posted links that "prove" the circuit you draw is called a series-parallel combination, you insist it is parallel, but change that ever so slightly, and now it's down to just personal comments and trying to catch me on choice of words. Give me your credentials before we continue or shall we do a test? I only react once to posters who make negative personal comments about me and this is that post.

I hope my posts help other reading them.

ciao!
Nick.

[MidLandOne]

No, no I don't suffer from the affliction of being offended at what people say - I do get amused though .

[Extemporaneous]

Originally Posted by MidLandOne

I'll expand out my last post which you seem to have ignored the possibility of its being correct (I sometimes wonder if the effort is worth it ).

MidLand,
I most certainly didn't ignore your last post. I guess I was typing and did see that you had rung in. Can't say if it's worth it to take the time to try to explain to others what they don't understand, sometime I'm sure its not. I can say that I appreciate it. Thank-You.

Originally Posted by chala

Extemp.
I had a look.

Thanks chala.

Originally Posted by chala

The bad part.
In regard to the first disclaimer I would not buy.

You mean the one that is in the Manual (see attached) that you generally only get after you buy the product?
Quit different from the statement they make that is on their Website (the one you generally see before you buy)(also see attached).
I'm sure it will be a good piece of equipment.

Originally Posted by s/v Jedi

Hmmm... my post must have gone by unnoticed or incomprehensible.

One of those two.

Originally Posted by s/v Jedi

OMG I hope this makes sense for non-electricians ;-)
cheers,
Nick.

Nick,
I hope you don't mind if I take a bit of time to confirm your theories. (Kidding!). This has been VERY Interesting!

Originally Posted by s/v Jedi

I did put quit some effort in it too.
Nick.

Clearly you put much effort and it (all the info in this thread) has been very interesting. I'll have to read it many more time before I can call it Educational. Educational would presume that I learned something. To learn something the presumption is understanding. I'm not quit there yet.

Originally Posted by s/v Jedi

I hope my posts help other reading them.

ciao!
Nick.

If you guys could keep this dialog going, I bet it would help a lot more!

Thanks to all.

Extemp.

[Therapy]

Midland, I am afraid Jedi is right.

So, who makes the best isolation transformer.

best-------> worst

expensive----> cheap

[s/v Jedi]

Extemp, thanks you. I would suggest you post questions or ask about what part isn't clear. I am sure I can explain things more clearly.

I did notice one thing you wrote: when the primary side is configured for 230V and thus there is one long series-circuit of LIVE -> FUSE -> WINDING-1 -> WINDING-2 -> FUSE -> NEUTRAL, you wrote that you think it's okay to have the breakers in series and both will break the circuit. That is unlikely actually. It is very difficult for manufacturers to make these thermal breakers within close tolerances... differences will be there. This means that one of them will go first. This breaks the whole series-circuit and the other breaker will probably cool down without triggering. So, in most cases, I would expect to find only one of them tripped.

But in 120V input mode, both will activate. One will go first and this takes the winding it is protecting off-line. But the load connected to the secondary is still there and there will be a surge in the 2nd winding (it wants to double it's power-transfer to the electro-magnetic field) and the 2nd breaker will activate to prevent that.

About in-rush current. This happens because at the moment of switching power to the transformer, it doesn't have it's electro-magnetic field yet. The impedance of the primary windings at that moment is equal to the DC resistance of the winding, which is very small. As the magnetic field builds up, the "impedance" (=AC resistance) of the wire increases and the current decreases to almost nothing. The current still there afterwards (without loading the secondary) is to maintain that electro-magnetic field.

The problem with this "start-up surge" is that it is often too big and breakers will flip. There is a simple solution to prevent that: limit the current during startup. This is done by inserting a resistor in series with the winding. But after startup, we don't want it anymore. This is solved by a relay that shorts the resistor. You can see these components on the circuit board of the Victron. Each primary winding has two parallel switched (can see the circuit-board trace) big resistors, each 47 Ohms 5W max., connected in series. The two parallel resistors equal a single 23.5 Ohm resistor that can dissipate 10W. This limits the inrush current to 120/23.5 = just above 5A. The resistors are mounted up from the board for better cooling. The relays that short them out after start-up are next to the breakers. In the upper right part of the PCB is an 8-pin IC. I would almost bet it is a micro controller (PIC or CTIC; there's a ceramic resonator to the right of it) which control & times the circuit and closes the relays. Older designs would have a timer-chip like the NE555 and older still would have old fashioned designs with relays-diodes-capacitor circuits for timing. I think this is a top and modern unit!

Well, let's analyze the rest too:

There are also relays connected in series with the secondary windings. These are closed at the end of the initializing sequence, when the output voltage has become stable. They are the ones on the right side of the PCB. This means that during startup, you will hear relays clicking twice before you get power on the output.

The only question left is the part on the middle bottom section. Part of that is a power-supply for the fan but there's more than I can't identify.

Something I saved for last are the two big blue blocks. These are transformers. Their primary windings are parallel to the primary windings of the big transformer. They have two functions: first, power the circuits on the PCB without having to deal with the surges of the big transformer, and also to detect and measure voltages applied to the transformer windings. That last part ISN'T USED YET. Look at the terminals for input and output wiring: there is room for more relays there. Victron will sell a version that automatically configures the windings for the choosen output-voltage! They only need to change the firmware in the controller and pop those relays in.

Last thing I detect is that you need to put a jumper in to connect neutral-out to the housing (=boat ground). That should be default/standard imho.

Considering it all, I like what I see and would have designed it the same myself if for myself. It's a Dutch product like myself though, so probably designed by one of my old classmates ha ha!

ciao!
Nick.

[s/v Jedi]

Originally Posted by Therapy

Midland, I am afraid Jedi is right.

Therapy thanks... Jedi's are noble knights and always right!

So, who makes the best isolation transformer.

I don't think higher price will always mean the better unit. Important features are:

1. stainless steel or aluminium housing
2. terminals for 3-wire input and 3-wire output. The input-ground should be isolated from the output-ground (sounds obvious but my Charles transformer didn't do that!)
3. a shield around the primary windings that is connected to the input-ground.
4. the outer housing connected to output-ground.
5. big schematic of jumper configurations on housing

Nice extra's would be rotary switches for configuration instead of jumpers, output voltmeter and a ground-fault-protection-device on the output.

The output-neutral should be connected to the output-ground but that can be added easily.

cheers,
Nick.

[chala]

Hellosailor
Glad to hear that some people are smarter than us. Here people need to be protected from them-selves, anyway that’s what the government reckons.
Communication between people is all about understanding and flexibility. I would hate to hear that, due to a misunderstanding, somebody gets hurt. Even a world top organization that confuses meters with feet can crash a mission.
Remember Fred and Ginger “you say…, I say”
Petrol, kero and paraffin could be the same thing depending where you are.
I am afraid I do not own an EnglishYnGlitch dictionary but I posses others and, for instance, an isolator in one language is called an insulator in another language.
I live in OZ and to the best of my ability I try to pass on what I know to help others. It is only advice (see Encarta for definition) and points of view that do not have to be followed.
Thanks for the lecture on superconductors. When I started selenium rectifiers were the flavour of the month.

Nick.
This is top stuff; you are a courageous man. Now you can explain to all PF, W, VA and why mirage fighter planes used to have 400Hz on board. Just joking.
Now just one remark:
In your post 40 it read “if you short a couple of turns”. I think this should read “remove a couple of turns”. People may take the word short as “short circuit a couple of turns” and may think that short circuiting (bridging a coil) is the way to go for changing voltage. Correct me if I am wrong.

John M. NADON and BERT J. GELMINE M. S. University of Detroit (US) in their excellent book INDUSTRIAL ELECTRICITY write on page 319 “a simple transformer may consist of two coils…” so this term seems to be correct in the US.

[s/v Jedi]

Chala,

PF is easy... only buy stuff with PF=1.0 After that, VA equals W so all trouble avoided ;-)

I didn't even know jet fighters had AC aboard and I can't believe they have transformers (weight). But the higher the freq., the smaller the components to convert voltage. This is also true for transformers.

shorting turns: while I wouldn't advise cruisers to do it, it is actually an accepted method. If you have 2 turns besides each other, remove their insulation and put a drop of solder on, you remove 1 turn of the winding. The problem with big transformers is that this drop might melt because of the high power involved. I have seen shorted turns in brand new mass-produced products like audio equipment. For big transformers, taking turns off becomes easier, but a lot depends on the way the transformer is constructed. I have seen big transformers prepared for shorting turns by attaching clamps on them. My Charles transformer aboard has taps on the primary windings, a couple of turns from one end of the winding. This would be the preferred method.

ciao!
Nick.

[Extemporaneous]

Now your just showing off!

In all seriousness, Thanks.

It will be a bit of time before I have any questions for you. I'm going to have to work through this first. It's all well beyond any of my electrical experience.
At this point my most intelligent question might be...........
"Pardon" .
Perhaps it's not that bad, but I will have some reading to do.
I'm NOT an Electrician.
I will work on it through.
Thanks again.

Extemp.

Originally Posted by s/v Jedi

Extemp, thanks you. I would suggest you post questions or ask about what part isn't clear. I am sure I can explain things more clearly.

I did notice one thing you wrote: when the primary side is configured for 230V and thus there is one long series-circuit of LIVE -> FUSE -> WINDING-1 -> WINDING-2 -> FUSE -> NEUTRAL, you wrote that you think it's okay to have the breakers in series and both will break the circuit. That is unlikely actually. It is very difficult for manufacturers to make these thermal breakers within close tolerances... differences will be there. This means that one of them will go first. This breaks the whole series-circuit and the other breaker will probably cool down without triggering. So, in most cases, I would expect to find only one of them tripped.

But in 120V input mode, both will activate. One will go first and this takes the winding it is protecting off-line. But the load connected to the secondary is still there and there will be a surge in the 2nd winding (it wants to double it's power-transfer to the electro-magnetic field) and the 2nd breaker will activate to prevent that.

About in-rush current. This happens because at the moment of switching power to the transformer, it doesn't have it's electro-magnetic field yet. The impedance of the primary windings at that moment is equal to the DC resistance of the winding, which is very small. As the magnetic field builds up, the "impedance" (=AC resistance) of the wire increases and the current decreases to almost nothing. The current still there afterwards (without loading the secondary) is to maintain that electro-magnetic field.

The problem with this "start-up surge" is that it is often too big and breakers will flip. There is a simple solution to prevent that: limit the current during startup. This is done by inserting a resistor in series with the winding. But after startup, we don't want it anymore. This is solved by a relay that shorts the resistor. You can see these components on the circuit board of the Victron. Each primary winding has two parallel switched (can see the circuit-board trace) big resistors, each 47 Ohms 5W max., connected in series. The two parallel resistors equal a single 23.5 Ohm resistor that can dissipate 10W. This limits the inrush current to 120/23.5 = just above 5A. The resistors are mounted up from the board for better cooling. The relays that short them out after start-up are next to the breakers. In the upper right part of the PCB is an 8-pin IC. I would almost bet it is a micro controller (PIC or CTIC; there's a ceramic resonator to the right of it) which control & times the circuit and closes the relays. Older designs would have a timer-chip like the NE555 and older still would have old fashioned designs with relays-diodes-capacitor circuits for timing. I think this is a top and modern unit!

Well, let's analyze the rest too:

There are also relays connected in series with the secondary windings. These are closed at the end of the initializing sequence, when the output voltage has become stable. They are the ones on the right side of the PCB. This means that during startup, you will hear relays clicking twice before you get power on the output.

The only question left is the part on the middle bottom section. Part of that is a power-supply for the fan but there's more than I can't identify.

Something I saved for last are the two big blue blocks. These are transformers. Their primary windings are parallel to the primary windings of the big transformer. They have two functions: first, power the circuits on the PCB without having to deal with the surges of the big transformer, and also to detect and measure voltages applied to the transformer windings. That last part ISN'T USED YET. Look at the terminals for input and output wiring: there is room for more relays there. Victron will sell a version that automatically configures the windings for the choosen output-voltage! They only need to change the firmware in the controller and pop those relays in.

Last thing I detect is that you need to put a jumper in to connect neutral-out to the housing (=boat ground). That should be default/standard imho.

Considering it all, I like what I see and would have designed it the same myself if for myself. It's a Dutch product like myself though, so probably designed by one of my old classmates ha ha!

ciao!
Nick.

[hellosailor]

Chala-
You didn't buy that story about inches and meters getting confused, did you? So far 2/3 of the Mars landers/probes have gone black. Now, which is more plausible to you? That NASA doesn't know how to measure? Or...that the Martians shot 'em down?

I vote for the Martians! (G)

Even a confirmed cynic like myself can't believe that the cream of Earth's scientific establishment doesn't know how to spec numbers. Although, I give them credit for being way more original than "the dog ate the cover letter with the specs on it".

[Cowboy Sailer]

Jedi (Nick) that was an excellent explanation of a 4 coil transformer. I thought about jumping in but you said it all.

[NorthwestDragon]

Could I get a few educated comments on this article?

SmartGauge Electronics - Isolation Transformers or Galvanic Isolator? 1/2

[Cowboy Sailer]

I tend to agree that isolation transformers are better. The currents in galvanic corrosion can be very small. Diodes have some leakage in the reverse direction and are not perfect. If one also uses the transformer to convert ac power voltages, like from 250 v to 120v, it might be worth the much greater expense.

[s/v Jedi]

Originally Posted by NorthwestDragon

Could I get a few educated comments on this article?

SmartGauge Electronics - Isolation Transformers or Galvanic Isolator? 1/2

Well, I read the document and think it's either dated or old news. Any voltage on ground big enough to put the diodes in conduction (1.2V for most systems) defeats the protection. As the DC negative is always connected to AC ground, it counts too. Think about fridge compressors running (Danfoss controllers that make high frequency 3-phase AC out of DC), VHF/SSB radio's, windlass, electric winches, inverters, chargers etc. etc. all put some garbage on the DC negative and/or AC ground and/or AC neutral (which is interconnected too).

The statement about a voltage present on the ground lead from shore is very true and mostly the rule in the tropics. The galvanic isolator is defeated continuously under those conditions. I have seen more than 30V on the ground lead and wouldn't be surprised to see even more (but 1.2V is enough for defeating the isolator!!)

A good isolation transformer is three times the price of a galvanic isolator and this is a very good investment for both the boat and your health. I am impressed with the Victron transformers: a 3.6 kW version costs EURO 650 listprice now (but that would translate to $875 with the currently weak dollar). A galvanic isolator with it's questionable protection lists up to $300 which is a waste of money imo.

cheers,
Nick.