Proton BASIC Community

I have a transformer I would like to use for an app.
The size is 240mm x 240mm x 120mm.

I remember one could get the transformer sizes and the VA rating off the internet.  Off late you get the most ridiculous sh1t as result :-(

It would be awesome if someone has a table with core sizes and the corresponding VA somewhere...

The e-core would be split so the center core would be 120mm as well, thus 120mm x 120mm.
Any idea what size core this is ?  I estimate it to be 3000VA to 5000VA, most likely 3000VA.
I need an approximate so I know what size diode bridge will survive on it.

Thank you for looking...

And yes, it is really HEAVY.  I had to take extra vitamins TWICE just to lift it on the work bench.
I want to add.  One thing I really like about this transformer is the neat way it was made.  Really heavy duty windings made for power.
It's like the new micro that can do twice the speed...

Edit, sorry wrong link:

https://www.giangrandi.ch/electronics/trafo/trafo.shtml

Based on transformers in my inventory, a safe estimate for your transformer is 2500VA.

Quote from: trastikata on Apr 04, 2025, 08:15 PMhttps://www.giangrandi.ch/electronics/trafo/trafo.shtml

According to the calculator the core area in mm is 120 x 120 = 14 400mm^2 and is a 10 KVA transformer
Core power: P = 1.096e+4

QuoteBased on transformers in my inventory, a safe estimate for your transformer is 2500VA.

Perhaps.  Could be right too.

For 120mm x 120mm core section, 2.5 KW is very little. I have 220v to 110v transformer with similar core rated 5 KW and in the same time I have 110v voltage stabilizer with much smaller core rated 10 KW ... 

Just as a rough indication I weighed the transformer, and no wonder I groaned when I lifted it on the workbench.  It weighs just over 50 kg.

Here's another thing.  Us guys get lower back problems when we lift heavy things, while our wives do not.  I'm now going to get her to move it around 

There is a way to determine the maximum current for a diode bridge for what ever the transformer can supply and that is to simply throw the baby out with the bath water.

I thought I had some 1200A 1200V diodes in a box which I stripped out of scrap telecommunication systems about 100 years ago, but they all turned out to be SCR's, and there are smaller ones too.  Can still do if I make a firing circuit to keep them on, but it's going to be a pain to do since you have to clamp the devices between two plates or they won't conduct.

I have some 25A square diode bridges.  I wonder if one can parallel 50 or so of them.  I hate these things, these 25A diode bridges already heat up at 5A.  I think they use poor quality silicone and get a huge voltage over each diode hence the heat.

And to find a shotkey diode...

Some feedback -
I made a diode bridge paralleling 8 of those pathetic diode bridges (25A 600V) and screwed them on a 6mm aluminum piece of channel I have.
The transformer is 220V - 45V, after rectifying you get 45 x 1.414 = ~64V

Connecting the DC to four 100Ah 48V batteries, the transformer happily supplied 80Amp, (20A to each battery) and then the mains tripped, exceeding 15 Amp of the circuit breaker.

I then used a big variac and adjusted so the mains will not trip and output to supply a total of 20A, ie 5A to each battery.  The battery packs regulate self which battery gets how much power. 

The primary concern adjusting the current so low was the temperature of these stupid diode bridges.
You can parallel them because the voltage across each diode increase with current, which will have the next diode conduct should there be small differences between them.
The total current of 8 x 25A diode bridges is supposed to be 200A, but they already got uncomfortably hot with the aluminum channel at only 20Amp (2A5 each).

Never the less, I went to bed at around 1h30 this morning.  The wife woke me around 8h30 with the news the batteries are at 100%.  This worked very well.

This I had to do because of the cloudy weather, we had less than 5% solar power most of the day, and the batteries were at around 60%.  It has been like this for a few days now.

The day before I had the generator run for a few hours, delivering 220V at about 9 or 10Amp, but that made little difference to the state of the batteries.  9Amp /4 batteries is only about 2A and to make up 40Ah the generator would have to run for about 20 hours.

In this the transformer is brilliant since it works on a power exchange basis, the current gain on the lower secondary makes all the difference.

Now someone has to just add wheels to the transformer because it is uncomfortably heavy (wife refuse to carry it for me), and to find a proper diode bridge.  To limit the primary (and secondary) current I'm going to use a series induction coil like some welder transformers have.  By adjusting the core in and out the output current also adjust, and will be more practical and robust than having to tote the variac around.

Sorry I want to add -  The generator (in case of no Excom) can only charge a maximum of say 9A.
The exchange ratio of the transformer is 220 / 45 = 4,9, lets say 4,5 for losses.  If I drive the primary of the transformer with 9A the secondary current would be 9A x 4,5 which would be about 40Amp, or 10A for each of the four batteries, a much more effective way to charge them.

The solar panels are awesome, even though they are only around 21% effective.  Even on normally cloudy days they produce power, sometimes less than full sun, but sometimes more than full sun, just depends how the sun is reflected by the clouds, up to 120 - 140%.

It's a pity you can't charge the batteries in series which would reduce the current and losses in your poor bridge rectifier(s).

So I would suggest using active rectifier(s) to greatly reduce your diode drop losses.

John

Hi John, the problem is not the diode drop losses as you put it, but rather the amount of heat generated as a result.  I've even had that if you solder wires on the tabs that they de-solder, the bridges get that hot, on a heatsink and well below the rated current.

I stumbled on this page, seems to be in Ukraine as the currency is UAH and the language is Russian.
https://asenergi.com/catalog/tiristory-silovye/ts171.html

On a sub page I found these
https://asenergi.com/catalog/tiristory-silovye/t123.html
Called tablet thyristors, who would have thought...

These are the same size as the small thyristors I have, so I assume the current will be roughly in that range, from 276 - 1122Amp and 250mA gate current and it seems 2V5 min ? to switch on.

I tested one yesterday and these don't require clamping to conduct.

If I assume the ones I have can only diode 276A, that is substantial, not even the total mains supply here can blow it up - that is what I call a safe device 

I have a low voltage power supply that can current limit 0 to 100A.  I think if I use that I can determine at what current the device begin to heat up.  For my application, if I don't need a large heatsink it would be inexpensive and would be easy to make a diode bridge from them.

The other challenge would of course be to make a firing circuit for it.  I will have to think how to do it.  Maybe I can finally get a pic worked into it somehow ;-)

Check STM electronics, they have 170 V, 200 A dual Power Schottky Rectifier for about 25$.

https://www.st.com/en/diodes-and-rectifiers/stps200170tv1.html#sample-buy

Thank you Trastikata, but you must remember that we don't have a currency in Sick Africa.  Anything in $ is 20 times more expensive, and that is even before Trumpet's exchange rate and sanctions 

I have the thyristors, if I can work out a sensible way to fire them it will work.
I'm currently working on getting the wheels done, this thing is bloody heavy !

Quote from: Fanie on Apr 06, 2025, 06:30 PMThank you Trastikata, but you must remember that we don't have a currency in Sick Africa.  Anything in $ is 20 times more expensive, and that is even before Trumpet's exchange rate and sanctions 

Digikey says they sell it in ZAR at 440 ZAR/pc with free delivery on orders to South Africa of ZAR 2000 or more..

https://www.digikey.co.za/en/products/detail/stmicroelectronics/STPS200170TV1/1007436

Quote from: Fanie on Apr 06, 2025, 06:30 PMThank you Trastikata, but you must remember that we don't have a currency in Sick Africa.  Anything in $ is 20 times more expensive, and that is even before Trumpet's exchange rate and sanctions 

I have the thyristors, if I can work out a sensible way to fire them it will work.
I'm currently working on getting the wheels done, this thing is bloody heavy !

Unfortunately SCR's have a forward voltage drop of between 1.5 and 2 volts so quite a bit of heat will still be produced, no better than a good silicon diode. Using FET's as active rectifiers with RDS(on) in the millohms region could reduce the heat well below what could be archived with the very best schottky diodes.

Using the SCR's that you already have could enable rectification and some sort of power control. This will depend on the application. A PIC and phase control could work OK for a resistive load.  Good luck with the project. BTW the Australian dollar has dropped below 60 us cents, it won't be long before my builds will be using old and recycled parts.

I have tested this Tim.  When the gate is constantly on, the device conducts fully at around 1,5V, which is about the same as most Si diodes.

So if you have 3 low voltage sources to keep the four devices switched on, they will act as a normal diode bridge would.

What would work very well as you suggested (and here comes the pic into play) is to control the power at the output by switching the devices, and so control the output power, much like a dimmer does.
Detect the zero cross from the AC, and adjust the on delay in the cycle to adjust the output power.

If I remember right, you need to control only two devices to achieve this.  The other two can remain on to act like diodes.

I do like the FET idea !  I will look into it.

The EPV191404 is an opto couple photo voltaic device.

I bet you haven't used these before.  We here have only one place selling them, don't know if you have anything similar, if you have and they are less expensive, I'd like to get some.

Turning on time is 0,2ms and off time is 0,05ms at 20mA LED current, not very fast, but switching may still be possible for 50Hz which is 20ms.

If you drive 10 to 20mA through it's LED it gives an isolated output of around 7V, suitable to switch most FET's.

I use them to switch my solar DC controller's FET's on and off, and the FET's can hence be in the high side or the low side.  The solar DC I switch are around 250V which drives 2000W oven elements.  Since I switch the FET's only on and off, there is no heat generation.
The FET's I use are Infineon's IPW60R070P6 50A 600V and 70mR RDSon, I last got them from Mouser.

The gate threshold voltage of the IPW60R070P6 is average 4V, the lower this voltage, the quicker the FET will turn on.  Low gate threshold voltage makes it possible for FET's to be used in low voltage applications or you will not get the FET to switch on properly, or not at all if the gate threshold voltage is higher than the used voltage.  A poorly switched FET is in linear mode and will better act as a heating device.

If you want to use the thyristors purely as rectifiers you don't need anything fancy to turn them on just two resistors per device but as already pointed out the dissipation is almost double that of diodes.  Diodes on a chunky heat sink with a cooling fan would be far better and the diodes used in welding applications are not expensive.

I like the MOSFET idea thought I never tried it myself.

Just been in the garage looking for something else and spotted this pair of dual IGBT modules that were removed from new dynamic brake equipment that was scrapped. 1200V 300A per device could be just what you need?

Quote from: See_Mos on Apr 08, 2025, 04:29 PMJust been in the garage looking for something else and spotted this pair of dual IGBT modules that were removed from new dynamic brake equipment that was scrapped. 1200V 300A per device could be just what you need? 

Personally I dislike the IGBT's, never got them to work properly.  FET's are far superior.

It is unbelievable how comfortable the large equipment handle power, I have a 4kVA variac driving the 3 to 5kVA transformer which both runs ice cold even under load.

I have ordered some 70A diodes, I'll parallel in pairs for a bridge and see if they live up to anything.

I havent tested the SCR's for heat at current yet, and I haven't had time for the FET bridge either.

Waste not, want not.
Heatsink the diodes to your hot water heater.