Maximizing Heat Dissipation in High-Power SMPS Design

[Salvador]

Hi, I am building this smps (schematic attached ) I have a question, since it's a half bridge topology the other end of the transformer primary is attached to half the mains rectified (capacitor middle) via a capacitor ,
but here's the question , the transformer primary is attached via a capacitor to the middle point of the smoothing capacitors which makes now two capacitors in series to ground and to +ve, if I am correct that makes the maximum voltage attainable at the primary not half but 1/4 that of the mains rectified correct?

can I just connect the float from the primary directly inbetween the smoothing capacitors without the additional capacitor ?
my devices are IRFP 460 they handle up to 400v if i remember correctly.

 

[Baluncore]

can I just connect the float from the primary directly inbetween the smoothing capacitors without the additional capacitor

No. The series capacitor 1uF/480V is necessary to make sure the current through the primary is balanced and averages zero.
If you connected the primary directly to the float terminal, the cap would need to move to the other side of the primary.
The capacitor also limits the energy transferred on each cycle of the conversion. It prevents high DC currents in the transformer primary.

 

[meBigGuy]

I'm trying to understand this.
You are saying the transformer needs to be AC coupled, which makes sense.
Is the problem being solved that when the primary current doesn't decay to 0 (fully discharge) because of rapid switching, a DC current will build up. The DC current would have to flow through the 27K resistors which would unbalance the switching point.

Or are there different/additional issues.

 

[Averagesupernova]

The directly rectified mains that is divided up with that capacitor divider can be thought of as a fairly high impedance source at DC but a very very low impedance source at the switching frequency. A fairly clever way of doing it I guess when there is no transformer to use the center tap for this float node. In a perfect world I think the series capacitor in question could be eliminated but I would not do it. Slight differences in tolerances could cause an imbalance in that capacitor divider. The series capacitor will not cause any loss. It is only responsible for passing AC and at the switching frequency it will not even appear to be there.

 

[Salvador]

Ok thanks folks for the responses , I think i got your point , will keep the capacitor , i have a 2.2uF/275V one , but since the voltage there is only half mains rectified i think it should do fine.

 

[Salvador]

also I wanted to ask one more question , see in the attached schematic there is a feedback optoregulator " 4n25" now it has two zener diodes in series connected to it , I assume the zener voltage is choosen to be such that its about 4 volts below the maximum output of the rectified high frequency AC coming from the transformer?

Also can I use more than two zener diodes in series but each being a different zener voltage ?
I just need to get some particular voltage but I don't have the correct zeners at hand but I have other ones I could combine hem in series to get the needed one?

 

[Baluncore]

Firstly, the capacitor. You should stick with a 1uF bipolar capacitor. Reducing the voltage spec might cause a problem when switching the unit on or off. There is a reason why they specified those values. It needs to be a good low loss cap, not electrolytic. The ripple current is quite high.

Secondly, the chain of zenners. You can make the chain from what you have but check them before you turn it on.
The 4 volts is needed to allow for the opto-coupler LED voltage, plus takes into account the difference between zenner knee voltage and at what voltage the zenner will be conducting enough current to turn on the LED in the opto-coupler.

 

[Salvador]

yes I found a wima labelled 1000/100 which I assume is meant 1000pF which is 1nf.
I put in a small potentiometer on the lead 6 of sg3525 to be able to change the frequency if necessary.

I only wonder what happens if I need say +-80 volts instead of 70, should I then put the zeners such that the voltage is about 76 volts to leave those 4v for the optocoupler led?

 

[Baluncore]

Yes, add up your zenners so they are 4V below the wanted output value.
Once you have it going you might adjust the combination slightly to bring it closer to the voltage required.

 

[Salvador]

so basically what your saying is that the sg3525 together with the octocoupler connected to the error amplifier in the Sg can work like a precise voltage regulator on the output, for example if I need 80 volts I put 76 volts on the zener series connection and the feedback changes either the duty cycle or frequency? and the output voltage hence changes correct?

 

[Salvador]

oh and one more question , is it okay that the driver mosfet ground is connected to the mains rectified ground but the grounds for both SG3525 and IR2110are connected to a separate small transformer rectified output ground?
see the attached schematic?

i have left all the small transformer ground with a black triangle and pointed out the mains rectified ground with red stripes as ground.

 

[Baluncore]

for example if I need 80 volts I put 76 volts on the zener series connection and the feedback changes either the duty cycle or frequency? and the output voltage hence changes correct?

Yes. But it is not precise, it is simple design, maybe +/-1V.

All small triangles are connected together as a common reference. They are all connected to the negative of the rectified AC.
The small transformer and 7812 regulator provides a 12V DC supply on the neg rectified AC input. It is needed to start and run the converter.

You must be careful testing that circuit. I use a 1:1 mains isolation transformer and Earth the common reference when working inside live switching power supplies.

Notice that the transformers and opto-isolator form an island of floating supply, often with 350V DC and capable of killing. See attached.

 

[Salvador]

well my question was about whether i can connect together the ground from the mains bridge rectifier and also the ground from the small rectifier that i have after the small transformer which feeds the IC's ?
the small transformer is made with a rectifier a smoothing capacitor and a zener which gives me about 24 volts DC that I feed into the sg3525 and ir2110, the maximum allowed for the sg is 35 volts and for the ir2110 25volts.
but should i connect the ground from the low voltage bridge rectifier together with the one coming from the mains rectifier?
I ask this because they are both separate units as we know and can't there arise some ground currents or unbalances etc?

is it possible to also keep the separately connected the mains for the driver transistors and the small ground for the IC's or isn't that possible because the IR2110 driving the mosfet needs both the +ve and ground as a reference to drive the gate properly or what?

thanks.

 

[Baluncore]

well my question was about whether i can connect together the ground from the mains bridge rectifier and also the ground from the small rectifier that i have after the small transformer which feeds the IC's ?

All triangles inside the red box MUST be connected together. You must stop calling those triangles ground. Those triangles are all a common reference rail at the negative rectified AC voltage. They must not be connected to anything else as they are referenced to Earth for RF noise only through the 2.2nF/250V capacitor. For DC and power they are referenced to the rectified mains AC through the 'float' circuit.

You should not be working on what is called "live chassis" mains circuits without more experience. You are really beginning to worry me.
Live chassis circuits have only one aim in their life, which is to take your life. It will try to kill you when you least expect it.

 

[Baluncore]

The circuit attached here should make more sense.

 

[Salvador]

I know everyone here on PF always insists on some overly worried safety advice , but don't worry Baluncore , I have been working with electronics for some years now, just because I'm not a pro at SMPS and some other stuff doesn't mena I don't know were the mains voltage is and were it's not.I actually always take great caution when doing things around mains simply because as a kid I got zapped by mains many times and it was out of curiosity back then I had a passion for those things back then but didin't had the respect towards them that I have now in my early twenties.

also I want to say thanks for your help.as I said I asked this about the grounds beacuse I the situation confused me a little because the mains has just a rectifier but the low voltage IC supply has a transformer feeding them so they are basically sort of only magnetically coupled to the mains versus the rest of the circuit high voltage side is directly coupled to the mains via the bridge.

but still I assume the load current from the transistors will all flow towards the mains rectifier negative not the small rectifier because there the current would have no where to flow since it's not directly coupled to mains correct? In other words connecting them together posesses no danger to the IC's ?

Oh and also when I power it up I guess it would be better to always have some kind of a load attached to the outputs from the secondary side.?

 

[Salvador]

now I want to share something , i put the whole thing together and it worked ... worked in the past tense is correct.i attached a 70w 230 volt bulb to the outoput and it worked the bulb was glowing and i got that nice feeling inside.
then a few hours later i thought to attach a 500w bulb as he load just to see does the voltage sag under load , as i connected the plug into the wall socket at first there was nothing i guess maybe i didint fully plug it in but after the second try a second split later all i heard was a loud bang and now one of the mosfets is blown open and the IR2110 is alos cracked open ,
either i had an unlucky pluggin in event as to maybe i kinda had contact for a little while then no contact and then contact again which could have caused some spike in the transformer or something or either these simple design smps are no good for any serious amount of load.

still i guess nothing beats the old but heavy and otherwise bulky transformer. :(

also the feedback optocoupler wasnt quite doing its job or maybe somethign went wrong there the first time i measured with the 70w bulb the output voltage was +-70 or 140v DC.the zener i had before the coupler was rated 82 volts.
maybe the octocoupler did some bad stuff when the mosfets blew the second time it senced low voltage or what and maybe turned the frequency or duty cycle for the sg3525 up which could have also contributed the blown device outcome?

 

[Averagesupernova]

...because the mains has just a rectifier but the low voltage IC supply has a transformer feeding them so they are basically sort of only magnetically coupled to the mains versus the rest of the circuit high voltage side is directly coupled to the mains via the bridge.

Ok, this is why you are getting safety warnings. Just because the IC is fed power through the low voltage transformer does not mean that it is isolated from the line voltage. The actual + and - 70 volt output is isolated but that is where it ends. Did you not look at the diagrams baluncore posted with the red line drawn around?

 

[Baluncore]

You now know what happens to SMPSs when they are being tested. What can I say? You attribute failure of the prototype to bad luck. The truth is that you do not understand the importance of knowing the circuit and getting everything right before applying power. For example, how did you calculate the area of the heat-sink needed on the MOSFETS? Did you check the temperature of the components with an IR thermometer before increasing the load? Did you use polycarbonate, ceramic or electrolytic capacitors in appropriate places? Did you Earth the common output or did that voltage float far enough from Earth to break down the PCB insulation?

still i guess nothing beats the old but heavy and otherwise bulky transformer. :(

Switching converters beat heavy mains transformers almost every time. Why? Because the switching frequency is one thousand times greater than the mains frequency, so the transformer can be much smaller and lighter.

but still I assume the load current from the transistors will all flow towards the mains rectifier negative not the small rectifier because there the current would have no where to flow since it's not directly coupled to mains correct? In other words connecting them together posesses no danger to the IC's ?

Have you looked at the “Modified smps 2.png” I attached to post #15. The 7812 regulated supply is a quite independent and isolated 12V supply that is only needed to power the SMPS chips on the –170V rail of the rectified AC. It is needed to power the SMPS chips before the switching supply can start.

 

[Salvador]

Ok, I am getting a fair amount of criticism some fair and some unfair.Heres what I have done. I really take care about mains and line voltages , after all I am soldering my stuff with a home made iron that I wound myself and it is being supplied about 40 volts from a transformer so it's decoupled from mains , by the way I did this because commercial budget and cheap soldering irons after a fair amount of use and time tend to zap you so I got this done , used the very old case of the mains powered soldering iron just rewound the coil and put some new isolation in.

as for the smps , the heatsink is actually better than needed , I used an old CPU heatsink from a dell computer drilled some holes in it and attached the transistors.Its a copper sink and i had a thermometer attached all the time when I was working with it it never got over 30 celsius, but even without looking i felt it with my hand and it was room temp so heat isn't an issue here. (yes the transistors are isolated from the sink just in case you ask)
after all Baluncore if you were reading what I wrote I had no way of checking the temp after the increase in load because while attaching the 500w bulb , obviously i turned the power off , and then turned it on again as would seem logical. the blast was so fast no one could have checked any temps.

yes I used standard polyester little boxes were a unipolar cap was asked and electrolytics were a sign was noted also ofcourse electrolytics for the mains rectifier smoothing.

yes I looked at your diagram.I also agree that in terms of space and weight an SMPS beats an oridnary transformer and ofcourse it's because of the switching frequency.although we could still debate whether they are a universal " one size fits all" solution to everything.

well i guess will have to sort out the problem with some thinking reading trial and error alltogether.I mainly think it has to do with the transformer as it is (core material windings etc) not matching the transistors under certain conditions , like heavy load , I mean after all the schematic is not that hard it's either back EMF that killed the setup or shoot through. Or the load was simply too big for the IRFP460 although i doubt that since they are rated Idc 18 amps at max 400 volts. but who knows.
sadly i don't have a 200w bulb should have started with gradual increase in load.

please can you comment anything about the condition why the devices failed after all the things I have said about how the setup is built? can a shoot through happen in this situation shouldn't the IR2110 have some built in circuit that senses such a condition in pin5 and alters the firing rate of the mosfets , the lower one at that point, or am I expecting too much from a simple single half bridge driver IC?

I have attached a circuit I found online , it seems someone has put a small circuit that senses shootthrough and shutdowns the oscillator IC, or maybe I am understanding it wrong.

 

[Baluncore]

although we could still debate whether they are a universal " one size fits all" solution to everything.

"Universal" means a SMPS can receive power from a range of grid voltages of 100VAC to 250VAC without a need for switching.

We have no idea how your SMPS circuit is built, so we cannot be expected to devine where the problem may have started.

Adding circuit complexity to trap and protect from fears will lead to lower reliability. Protection and reliability are gained by getting it right the first time, then not cutting corners in manufacture.

 

[Baluncore]

Instead of IXFH32N50, you used IRFP460 for your MOSFET switches.

Internal diode worst case reverse recovery time went up from 250ns to 860ns.
Continuous drain current went down from 32A to 20A.
Rds(on) went up from 0.16 to 0.27 ohm.
And that is just the start …

You should have de-rated power and switching frequency when selecting a slower cheaper mosfet.
It is not really surprising that the mosfet blew out and took the IR2110 with it. Cutting corners usually costs more than it saves.

 

[Salvador]

I wasnt cutting corners as much as I was just testing the design at first with what I had at hand. Actually I am not so saddened about the fact they blew up I made everything with a thought in mind as if something blows up it can be easily accesed and changed.Yes I agree the IRFP460 are lower power devices.

Also I measured the bulb I attached when it blew up and first time i was measuring its resistance through the wire leads into the lamp turns out they were corroded and showed about 250 ohms which kinda puzzled me as it's not a very large load for this smps but then i took the bulb out today and measured across the bulb and the 230v 500w bulb measures actually just 10 ohms across it, so I think it's hardly a surprise why the MOSFETS went crazy.
I guess I should get higher rated mosfets for this application , even though the bulb was just a stress test and the initial goal is to power an amplifier still its better to have a reserve.

I'll check if I can get the original mosfets shown in the diagram the ones to whom you referred , in case I can't maybe you have some advice for some other ones I could use?

thanks.

 

[Baluncore]

The resistance of a tungsten filament is proportional to absolute temperature.
At room temperature = 300K, when incandescent it will be above 3000K.
If 10 ohms at room temp then it will be over 100 ohms when in use as a light.
The unit being tested must be able to handle the cold filament resistance.

 

[Salvador]

obviously this unit wasnt able to handle that resistance.If my ohms law calculation memory is still correct then at 10 ohms resistance and 140volts DC the starting current at that resistance was 14 amps , which corresponds to a starting power of nearly 2KW.

Now I did some search in the local stores provided catalog of devices and basically these two are the ones I could use.
IXFH24N50 and
IXTH24N50L

they are kinda expensive but not as expensive as the originally mentioned ones yet are very close or even similar in capability to the original ones showed in the schematic.

 

[Baluncore]

You seem to enjoy courting disaster. Instead, select a device that exceeds the original specifications.
The IXFH24N50 are only; 24A not 32A, 230mR not 150mR, 300W not 360W.

I would consider a more recent device with better specs and at a lower price than the poor choices.
For example; Fairchild FCH104N60. 600V, 37A, 104mR, 357W. It is less expensive than the IXTH24N50.

 

[Salvador]

thanks for giving me an example I will run it through the databases of some of my local shops and see whether they haev it or not, I agree that my previous examples were not the best choices it's just that sometimes local distributors ask a cosmological scale price for a device which is not that worthy or even obsolete. like the IXFH32N50 mentione din the schematic. one supplier even asked about 24 euros a piece which is about 27 dollars a piece, the other store had the same device at exactly half the price.

Ok I will check out the fairchild device.

 

[Baluncore]

Take a look at these global suppliers.
FCH104N60 digikey.com have 480 in stock at AU$8.35
FCH104N60 mouser.com have 378 in stock at AU$11.06

 

[Salvador]

I don't know about digikey but I just registered at mouser and yes the FCH104N60 itself was cheap about 6.3 euros but then I got to the shipping part and shipping alone costs 40 euros so I kinda thought no thanks.

the best I came up with now is to buy FCH35N60 from US through ebay , out of the many hong kong fake picture ads there seems to be a decent guy from the US selling electronics with a good track record. He offers 4 of these FCH35N60 for about 12 USD which is a good price as even at the l;ocal stores I usually can get thr standard IRFP's for such a price.
Are you from Australia ? I am from northern europe by the way , Latvia to be exact.

Also maybe I could just build a smart little circuit which could use the transformer as a magnetic amplifier at startup , to limit the current sort of failsafe soft satrt circuit, just say I added a small extra winding on the core. at powerup a capacitor would get charged up through this winding say the capacitor would charge for 1 second.as it would charge DC current would be flowing through the winding altering the cores saturation limit so that only half or even less of the AC could get through the core this would limit the current that could be drawn by the secondary at heavy startups if a 500w bulb is attached as load or anything else demanding, what do you think could it save an otherwise decent device from blowing up?
probably the IRFP 460 would have somehow driven the bulb i attached given that it would have been soft started or preheated.

still I doubt I would ever need this for the purposes on which I intent to use this smps ,

 

[Baluncore]

what do you think could it save an otherwise decent device from blowing up?

Building a complexity to prevent a fear is not productive. You are better to use a number of lower rated globes that can be switched in as needed.

Yes I'm in Australia. Take a look here ...
http://lv.rsdelivers.com/product/fa...0-v-3-pin-to-247/8651271.aspx?query=FCH104N60

 

[Salvador]

oh thanks for showing me this site , looks really good, I even checked up and started to order to see the shipping prices and they are reasonable atleast compared to what I saw elsewere, the problem is that I cannot finish the deal because it seems they are still developing their business and the homepage is temporarily closed so i cannot proceed beyind the input your data stage for odering , maybe will call them tomorrow to see what can be done.

 

[Salvador]

so i got the fch104n60 from the RS company 9by the way thanks for telling me about them)
Now I wonder what kind of a startup load i could use i mena the minimum for a test?
i kinda don't feel like putting the 500w bulb which had 10 ohms when cold. since the use for the smps will be in an amplifier it would rarely see such low resistance (typical speaker load would be 4 to 8 ohms but that's an impedance load not a purely resistive one and one rarely plays speakers to max)

i have a 230w 15 ohm bulb , i don't know what else I could use to see the output power of whatever i have built there?
or maybe i should take the 500w bulb and just use a reostat to slowly bring it to full power.
also I wonder does the soft start capacitor on the sg3525 acts only as a time delay depending on the capacitance used or does bigger capacitance actually startup the mosfet driving IC's slower? like more gradually.

 

[Baluncore]

Take the output and connect across it several filament globe sockets with switches. That enables you to switch in a variable load during your tests. You should select and design your load by using the cold resistance of the filaments.

Your +/–70V represents a 140V DC output. That is insufficient to fully light a 230V filament lamp. The cold resistance of a globe will be about 1/10th that of a bright globe. We know that because resistance is proportional to absolute temperature. Room temperature is 300K while an incandescent filament will be closer to 3000K.

I have 240VAC and use an equation to estimate Rcold = 3600 / Filament wattage. So a 100W filament globe will have a cold resistance of 36R. 75W will have 48R. Each switched filament will be a current load. The maximum current will be less than Vout / Rcold since the filaments will be warm, radiating IR. It will not be the cold or the hot resistance, it will be something in between. You can always measure the current and voltage to determine the filament temperature.

You can buy 230V filament globes from 15W to 400W. I suggest 15W, 25W, 40W, 60W along with several 100W globes. You do not have to have them all plugged in at the same time. During early testing you might progressively switch in low values, later during testing they might all be 100 watt globes. 400W will only be needed at the end during reliability testing.

For 140Vdc output it would suggest …
watt, Rcold, Imax
15W, 240R, 0.6mA
25W, 144R, 1A
40W, 90R, 1.5A
60W, 60R, 2.3A
100W, 36R, 3.9A
400W, 9R, 15.5A. Avoid the 400W filament initially. Go for lower values, then several 100W globes.

The other thing you can do is put a light globe in series with the AC input during early testing. It should flash on momentarily while the storage capacitors initially charge, then go dull once running. If it stays on, then the load is too great or you have a short in the AC switching side of your inverter. The series globe should minimise the damage from catastrophic failures that take out multiple components. For example, if your half bridge fails, or one mosfet goes short, then the current through the remaining mosfet will be limited to below the series filament globe specification. You will clearly see if you have a problem.

I wonder does the soft start capacitor on the sg3525 acts only as a time delay depending on the capacitance used or does bigger capacitance actually startup the mosfet driving IC's slower? like more gradually.

See this application note, page 5/13;
http://freedatasheets.com/downloads... Supply And Power Management SG3525 AN250.pdf

On page 6 is the equation for “turn on time”, t = Css * 2V / 50uA. With a bigger Css It will take longer to start.

 

[Salvador]

now here's what's happening as of lately between me playing football and repairing my car.
i changed the IRFP460 to the newly received FCH104n60, now take into consideration that just before the IRFP's were driven by the same board which has the SG3525 and IR2110 on it , a 115w 230 bulb glowed nicely and even though i didnt try higher powers with the IRFP'S yet i soldered in the FCH ones.
now the same 115w bulb attached that made no problems to the IRFP and guess what happens when i switch on the power with the fairchild devices , nothing , not even a crack or a bang , but the lights went out immediately, so as I went to replace my fuses, came back and measured and yep there it was , both devices blown open completely , not physically but electrically , shorts on all three legs G D and S.

everything else was the same as before i changed nothing except the two half bridge mosfets driving the transformer.Mystery.
All in all I have to say I have never been a fan of semiconductors in places were high power both volts and amps have to be dealt with as any solid copper wire and a piece of iron will take any abuse and tolerate almoust anything but a semiconductor will not even close.tahst what i hate , yes I know that almoust every household item is now powered through a SMPS and their reliable and so on but well ehh I just have no luck with them , maybe because for them to work properly they need a heavy and advacned circuit were one parts watches over the other and so on and so forth and everything is designed and built to the maximum precision.

the other option is that the devices I have been given are fake and hence can't stand up to the requirments , since the IRFP's worked in the same situation with being lower rating devices.

maybe this will sound a little childish but since it's all just a big experiment i guess i will stick with what works and buy myself some more IRFP'S and just see how much power such a psu can give to my amplifier board , both with the satisfaction of my ears and also a small oscilloscope amplifier output load test. if they will do no good well FFS , i'll just use a toroid , atleast that thing aint going to break down in a hundred years.

 

[Baluncore]

, both devices blown open completely , not physically but electrically , shorts on all three legs G D and S.

That is a contradiction. Open or short ?
Did you have a 100W filament globe in series with the AC when you turned it on ?

I have never been a fan of semiconductors in places were high power both volts and amps have to be dealt with

High current is OK, high voltage is OK. But never both at the same time.
That may be your problem.
There is a problem somewhere in your circuit. I have insufficient information to guess where.

the other option is that the devices I have been given are fake

That is most unlikely.