[contactirfu]

This thread shall discuss Power Supplies for UHU both HP and regular - and the regenerative dump circuits.

This thread will include posting of circuit schematics / calculations / diagrams and recomendations.

[contactirfu]

Hi Folks lets start this thread with my problems on the power supply end

First here is the list of parts going into the PS

1. 220V primary and 78V 20 AMPS secondary toroidal transformer (gives 110V after conversion to DC)
2. 50 AMPS Bridge mounted on Alu heatsink
3. 5x 5000uF 150V electrolytic caps connected in parallel to give 25000uF
4. 5x 1uF 400V film capacitors connected at the electrolytic capacitor terminals
5. 2x 1k 20W wire wound resistors - wired in series to get 2K 20 W (used as bleeder resistors)
6. Regenerative dump circuit (Henrik Design)
7. 3x Fuses 8 amps cartridge.

the problem:

When I connect the powersupply transformer to the mains, the phase fuse in the main's blows, I see the fuse has only 4 to 6 strands of thin copper wire in it. I replace it and then go in for thicker wire on the fuses. then i connect again and again the fuses in the main switch blow. again there are only four copper strands connecting them, I thicken them to like 10 to 15 strands, then I connect the PS again, now the fuse (8 amps cartridge) just before the HP UHU blows, and immediately the Bridge rectifier burns out.

This time the phase fuse in the electric pole outside is gone, the 32 amps MCB did not kick in!!!!

This is what I am assuming : the toroid has a high inrush current at the start: how much can that be? is that causing the fuses to burn out?

am I using wrong components anywhere?

what is the right way to check the PS step by step?

are there any extra components I need to add to the PS for safety concerns?

find the photos attached! this problem is giving me nightmares - kindly help me solve it.

RGDS
Irfan

[Khalid]

iRFAN..cHECK AT mECHMATE FORUM OR CALL pMINMO ...HE WILL HELP YOU....I hope the problem will be resolved by just getting outside and grab some electrician (not electrical engineer) who have many year of experience tackling such problems...
I think i saw somewhere Hight value capacitor being added at power supply...I don't know the exact reason...
Hope you will come up the problem soon, i saw the video machine is little slower with less voltages to servos...I hope you will get enough power to run at 200IPM...
Best of luck

[contactirfu]

Hi Khalid, thanks for the thumbs up bhai, I am asking all the people here ! whom so ever i can find. and also trying to draw a circuit of what I have ,

[Khalid]

[Al_The_Man]

Your problem is most certainly a defective component or mis-wiring, The normal off load inrush should not cause the effect you are seeing, even with Caps connected.
The recommended method of trouble shooting this kind of problem is to isolate sections at a time and power up as you go.
IOW, first power up the Toroid with the secondaries open, then connect the Bridge with caps out of circuit, most often you are going to find a bridge connected wrong/shorted or a wrong connection somewhere, the secret is methodical trouble shooting.
I haven't seen re-wireable fuses for decades.
The size of fuse you have there would supply a welder!
Al.

[contactirfu]

Your problem is most certainly a defective component or mis-wiring, The normal off load inrush should not cause the effect you are seeing, even with Caps connected.
The recommended method of trouble shooting this kind of problem is to isolate sections at a time and power up as you go.

Thats what I am going to do exactly! and follow a step by step procedure like you have said.

IOW, first power up the Toroid with the secondaries open, then connect the Bridge with caps out of circuit, most often you are going to find a bridge connected wrong/shorted or a wrong connection somewhere, the secret is methodical trouble shooting.

Have learnt it the hard way!

I haven't seen re-wireable fuses for decades.

That sort of things run most of the industrial (small ) shops here

and thats where we are.

I will replace all fuses with D-types once everything is running.

The size of fuse you have there would supply a welder!
Al.

I am going to have a stick welder soon too.

and yup........will do the PS installation step by step religiously this time.

Thanks for the inputs Al

[vroemm]

Hi contactirfu,

You have a few electric boards directly against the wooden plate.
If one of the boards gets hot because of a defect, the risk of fire is then bigger.

I would use distance tubes to make a distance between the boards and the wooden plate, which would reduce the risk of fire.
Although i would use a metal plate instead of wood in the first place.

Vroemm.
(Hmm, fun, writing in the new uhu forum :-) )

[vroemm]

5. 2x 1k 20W wire wound resistors - wired in series to get 2K 20 W (used as bleeder resistors)

2 x 1k 20Watt in series becomes 2k 40 Watts :-)

Vroemm.

[kreutz]

A 2 KOhm 40Watt resistor as a regenerative energy dump load is not going to be effective. As soon as I have some time, I will post the required calculation steps for the right resistor value. I would follow Al's procedure as the right troubleshooting procedure.

Kreutz.

[contactirfu]

Hi contactirfu,

You have a few electric boards directly against the wooden plate.
If one of the boards gets hot because of a defect, the risk of fire is then bigger.

I would use distance tubes to make a distance between the boards and the wooden plate, which would reduce the risk of fire.
Although i would use a metal plate instead of wood in the first place.

Vroemm.
(Hmm, fun, writing in the new uhu forum :-) )

The board behind is HYLAM a insulator regularly used in the electrical industry, any heating is localised and end residue is powder , however I will still use nylon standoffs for all, was just in a hurry to get everything on, I had concerns before you bcos as you can see that I have fixed the PCBoards using cable ties - funny you did not notice that one LOL, I am gonna take those things out! but not before all 3 axis are running.

( Hylam - http://www.bakelitehylam.com/indlamni.html )

The whole controller was built by me and my Dad, the whole Sourcing task done by me, and dad handling the logistics for me.

and YUP you are right on the resistor side - being a mechanical engineer has its disadvantages

Kreutz thanks for help on this one. hope you can make your calculations sooner, it will help most people out here.

RGDS
IRfan

[contactirfu]

Meanwhile one of the zone friend Manjeet suggest to me that If I use a switch between the Servo boards and the power supply after the capacitors then there will be an open secondary and I can have a delayed switching after the surge has stabilized, so I am going to have a DPST switch rated 20 amps in between the UHU servo power supply and the UHU boards. In future I will use a delay switch to switch on the PS for the UHU - hope it works!

[Gerald_D]

If you were using geckodrives, a switch between the capacitors and the gecko would be a very expensive idea.....goodbye gecko.

[contactirfu]

Thanks for updating me G' ,

meanwhile an electrical engineer friend and colleague gave me some horrific facts about my powersupply.

1. For the capacitance(25000uF 150V) I am using the initial surge current would go upto 1kiloamps.
2. The bridge bkpc 5010 has a surge current handling capacity of 400 amps, beyond which the bridge will blow, like it did in my case.
3. He calculated that the bridge can take upto 11000uF capacitance at 110V.
3. The capacitance might burst in use bcos is just 40V above supply voltage.
4.he wasn't sure why the 8 amps fuse blew near the UHU servo and asked me to run the motor again in the lower PS and confirm if mosfet have survived.

well he filled up a full sheet with calculations and I think he is correct, he wanted me to use some sort of pi filter which he would design with a new powersupply.

I have asked him to design supplies individually for each motor so that I can switch on each one individually? I donno if I am doing right or wrong.........

then I just mentioned to him that design the best economical solution and he wasn't sure if a single PS would be economical or separate ones?

What ever the case I am looking again at around 200USD (8000INR) + investment on this.

hope some one else has better ideas!

RGDS
Irfan

[H.O]

Hi Irfan,

What you need to do is limit the inrush current.

Attached is a schematic of a simple way of doing it that I think will work in your case.

The relay, K1, should be a 24V one with contacts rated at the expected maximum current on the primary side (10A will probably be fine). To calculate the value of R6 take your DC supply voltage, subtract the relay coil voltage and divide by the relay coil current.

Let's assume:
DC voltage: 110V
Relay voltage: 24V
Relay current: 55mA

110V - 24V / 0.055 = 1560ohm, (1.5k will be just fine)

The power rating of the resistor is calculated as (I^2 X R) so with the above numbers you'll get: 0.055*0.055*1500=4.5W (use atleast 10W)

So, how does it work.... When you power up the circuit the realy isn't energized so the current to the transformer need to go thru the 47ohm resistor, this limits the inrush current to around 5A. The voltage in the capacitor bank starts to build up and when it reaches a certain point the relay will energize and bypass the resistor, enabling the full power of the supply.

There are more advanced ways of doing this with timing relays etc but I think this should work in your case. If the relay you use is has two poles you could put a second resistor between the rectifier and the capacitors and bypass it with the other contact in the relay. Make sure it will handle the secondary current though.

/Henrik.

[contactirfu]

Henrick,

If i decide to put in a resistor in between the bridge and the capacitor bank, then what sould be its value?

RGDS
Irfan

[H.O]

Hi Irfan,
It's not at all critical, somewhere between 10-100ohm would do it. Remember, it's just used to limit the initial inrush current, when the capacitors are empty.

Use wirewound resistor of 25W or more. With a 47ohm resistor you'll probably get a delay of around 0.5-1s before the relay activates - that should be more than enough to limit the inrush current you're seeing.

Good luck!
/Henrik.

[contactirfu]

Thanks Henrick you have helped me to sound sleep today, will carry on the modifications tomorrow and update all.

RGDS
Irfan

[kreutz]

Meanwhile one of the zone friend Manjeet suggest to me that If I use a switch between the Servo boards and the power supply after the capacitors then there will be an open secondary and I can have a delayed switching after the surge has stabilized, so I am going to have a DPST switch rated 20 amps in between the UHU servo power supply and the UHU boards. In future I will use a delay switch to switch on the PS for the UHU - hope it works!

DON'T Do That!!!, it will ruin your HP UHU. Opening the switch will create a high voltage surge due to the inductance of the cables.

[vroemm]

What you need to do is limit the inrush current.
Attached is a schematic of a simple way of doing it that I think will work in your case.

You could add a simple way to unload the capacitors when the circuit is switched OFF.
I have not tried this circuit.

Works like described in previous message.
And when mains is switched ON, the relay K2 comes up.
Which disconnects the resistor with K2-P1.
When mains is switched OFF relay K2 comes down.
And connects resistor via K2-P1 to the capacitors.
The capacitors unload via the resistor.

Does this work ?
Or did i make a mistake somewhere ?

See the circuit image below.
The zip file also contains the eagle .sch circuit. (in case someone wants to improve it :-) )

Vroemm