Torroid vs. SMPS

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Eva:

Your Nokia phone charger, the first picture shows a "thick" yellowish wire, to me it looks like a "tripple insulated wire" which, if I got it right, use to be approved in acordance with the most common electrical safety standards, these type of wires is authorized as "Reinforced Insulation" and can be used directly on primary windings without any isolation.
For instance wire from Furukawa, (for a certain diameter and type as I have noticed), the dielectrical strength is upto around 25 kV.

Michael [/B][/QUOTE]

Yes, from my perception the pics of the Nokia charger did not look like poor isolation. Also they introduced cuts in the PCB to get the advantage of clearance vs. creepage.
The wire could definitely be the Furukawa TEX-E.

...25kV... well, the allowed dielectric strength is also a matter of frequency...
...coming poorer at higher frequencies. ...not only for the Furukawa wires, - this seems to affect most isolation materials.
 
This cut in the PCB of the battery charger was done on purpose. It has two small pointy electrodes (one on each side) spaced by 4mm, so an arc will jump if more than 5KV appear from primary to secondary side.


Nice choke, by the way. I have no access to litz wire nor triple insulated wire. Also, my core and bobbin selection is quite limited.

Actually, I had to look in German stores in order to find suitable stuff for my projects. I have bought cores and coils from Conrad Electronics, and heatsinks with mounting clips from Fischer.

I bet there is not much SMPS development in Spain.
 
50Hz transformers are not a good idea, too dangerous. A single transistor self-oscillating converter use old TV horizontal transformers (primary rewinded) could spark a several mm and output less than 1mA, powered by a single 1.2V battery.

The most easy way may be the spark thing in gas lighter, piezoelectric one could spark 5mm in free air, while big battery-powered one could achive more.

Eva said:
Also, I'm thinking about building my own insulation tester. I have some medium-size and big 50Hz iron-lamination cores that I can dismantle and rewind as 100:1, 200:1 or 500:1 step-up transformers (this time with gentle creepages and maybe some epoxy).

I could bet that those chinese transformers with no creepage doesn't break down until well above 5KV (there's no better way to know than testing).
 
kenshin:

I've been playing a bit with that neon lighting transformer that I found under a pile of stuff. It works fine and it's like new.

It sparks a bit more than 10mm between needles so no load voltage is probably between 10 and 15KV (obviously I can't use miltimeter here). Short-circuit current is 23mA. This is dangerous, it's just enough to shock somebody badly or kill him, but I'm pretty aware of the risks.

Note that these are special transformers with shunting rings on their cores that provide inherent current limiting, they work almost like current sources over a wide range of output voltages. I think that this feature is quite useful for insulation testing.

Bad news are that these units usually have their HV winding center-tapped to the chasis, but mine isnt :D , it's floating so I can referente one of the HV ends to ground in order to place a small resistor to measure leakage current.

I'll have to power it from lower voltages in order to get the desired no-load output voltage, though. Also I must get a high voltage probe.
 
About 50Hz transformers:

All small 50Hz EI transformers in my hand have primary & secondnary winded in seperated bobbin, "absolutly" safe creepage spacing. I have saw some big ones (>100W?) use unseperated bobbin with no side creepage, but they are very old things manufactured in 1970s.

I'm looking for seperated bobbin for SMPS but not found any, except common mode choke coils (something which is not for transformer use).

About PC SMPS:
I could see the wires in some PC SMPS main transformer from side, they have a distance ~ 2mm to the side of yellow tape. But PC is grounded equipment.

About phone charger & wires:
Some manufactures said on a Chinese SMPS design forum that their phone charger transformers passed 2.5KV hi-pot, the cost is 0.3 RMB (4 cents), it's a massive production price for >5 millions of products per year.

On the same forum I've heard some manufactures use triple insulation wire to wind their transformers. But it's relatively new things, not available to me. Could Teflon high temprature wire have the same safty voltage rating? These wires have more thicker plastic insulation layers.
 
'separated bobin' (split bobbin) transformers are not practical for SMPS since they show too high leakage inductances, neither they are practical for high VA 50Hz applications for the same reason.

Actually common-mode chokes are transformers, but leakage inductance is advantageous for these since it produces some degree of differential mode filtering. That's why split bobbins are employed.
 
About my SMPS project

I haven't begin any making work -- It's a project collaborating with my friend, he'll wind the transformer during vacation at home; and he could probably do vaccum potting in his father's factory.

I've bought two "100W" EI ferroite cores, my friend said that they are too small. He suggest to use a 200W core/bobin to wind a 100W transformer to leave enough spacing.

The secondnary will be winded with Teflon "high temprature" wires. The primary windings will be built thisway: put a layer of insulation tape per 2 to 3 layers of wires, leave 2mm creepage spacing on each side by folding the tapes back and block the wires (So the total creepage spacing is 4mm).

He said 10mm creepage spacing is too much, the biggest insulation risk is not breakdown to the secondnaries (almost impossible), but breakdown between primaries (this would possibly hapen and burn the SMPS). Besides, another friend said the quality is depended on building techniques instead of spacing.

Is 4mm creepage spacing + Teflon "high temprature wires"secondnary enough for equipment without grounding?
:confused:

The transformer will be test under 10KV after building, use a spark igniter.

About SMPS debugging: He suggest power the design directly from the mains: Wire up all probes and oscillocope before start, and stand away from anything when the system is plugged to mains.

ps:
Rather strange that you order cores directly from big companies. I bought these things from small shops in my city, HEFEI. It's the center of "the poorest" province in central China, and electronic things are not aboudant. Nearby province centers (Nanjing, Wuhan, Zhengzhou) have far more devices to buy.

EVA:
It's only half true that "Chinese companies are free to grow in their country, until they can afford approval costs in order to enter our market." Electronic companies also have to pass safty recognitions in China to sell their products.
 
Eva said:
This cut in the PCB of the battery charger was done on purpose. It has two small pointy electrodes (one on each side) spaced by 4mm, so an arc will jump if more than 5KV appear from primary to secondary side.


Nice choke, by the way. I have no access to litz wire nor triple insulated wire. Also, my core and bobbin selection is quite limited.

Actually, I had to look in German stores in order to find suitable stuff for my projects. I have bought cores and coils from Conrad Electronics, and heatsinks with mounting clips from Fischer.

I bet there is not much SMPS development in Spain.


spark gap:
:bigeyes: :bigeyes:
I had a closer look to your pic and indeed it looks like spark gap with electrodes on purpose. But why? I seems to jump prim/sec, more or less parallel to the transformer isolation.

My choke will have to show if it is just nice looking or really nice.
In fact the litz wire was banned from the winding lab of my former company, because we decided to switch to 155°C-wires and that litz is listed for 130° only.
If you go shopping in Germany don't miss out:
http://www.buerklin.com/default.asp?l=e
Their ferrite assortment is one of the best here in Munich.

SMPS vs. torroid...
Design/R&D is the interesting part of DIY. The building work itself
is often a little bit boring & annoying. Well, I have started these two Sadhara Cubes and already with my old electronics they sound fxxxxx good. :spin: :yes: :emoticon:
I decided to go ahead with my SMPS at least for one them.
If it's getting boring, I still can simplify the second with an torroid.
 
Why there's a spark gap between prim/sec, isn't it dangerous creepage? Or the designer expext use recoverable spark discharge in air to prevent permanent insulation breakdown during high voltage spike (thunder induction, etc. )?

ChocoHolic said:

spark gap:
I had a closer look to your pic and indeed it looks like spark gap with electrodes on purpose. But why? I seems to jump prim/sec, more or less parallel to the transformer isolation.
 
Kenshin said:
Why there's a spark gap between prim/sec, isn't it dangerous creepage? Or the designer expext use recoverable spark discharge in air to prevent permanent insulation breakdown during high voltage spike (thunder induction, etc. )?


Not creepage issue as there is hole in pcb. Clearance values apply.
I guess that if it really is desingned as sparkgap its purpose is to have controlled point of discharche, insulation break and surface flashover are much worser choices as the will/can lead to permanent degradiation in safety whereas airgap doesnt mind at all.
 
mzzj said:


Not creepage issue as there is hole in pcb. Clearance values apply.
I guess that if it really is desingned as sparkgap its purpose is to have controlled point of discharche, insulation break and surface flashover are much worser choices as the will/can lead to permanent degradiation in safety whereas airgap doesnt mind at all.


Hm, so it is preferred in case of an heavy spike on the mains to protect
transformer rather than the person? :hot:

: Ironic Mode OFF:
Or is the estimation as follows:
It is not probable that somebody touches the secondary right when
such critical event takes place. But surely somebody will touch sometimes after that and then it is better to have the isolation still working. ...would make sense for me...
 
ChocoHolic said:


Or is the estimation as follows:
It is not probable that somebody touches the secondary right when
such critical event takes place. But surely somebody will touch sometimes after that and then it is better to have the isolation still working. ...would make sense for me...

Or really powerfull ESD from user. I remember when i was kid that we were playing in winter at school, 25mm sparks were quite easy to generate on lacquered wooden floor when sliding with wool socks.
:D It hurted like quite badly on fingers so we kept fork or similar in hand to spread current and shock others with fork/spoon whatever :D

Too bad that after couple of years of fun they re-laquered that floor with some more modern stuff that was generating 5mm sparks at best. :D
 
If use a spark generator to hit something on the mains, will the spark sustain and conduct mains electricity to the generator?

It's a problem I often worry while playing with gas lighter ignitor: The spark of it is safe, but what would happen if use it to shock something on the mains?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.