Yes, why not to take a look over the ST's L6561 for example? It is more reliable that the Motorola (On Semiconductor) counterpart, almost in my experience repairing industrial electronics. Also the L4981 is like an industry standard.
Suggestions only.
Suggestions only.
Is the L6561 suitable for a 450W PFC? I only see small PFC on the reference schematics.
Yes, power capability depends in the MOSFET's current ratings. If it is suffiently high, no problem. At most you will need a MOSFET driver, like a bipolar emiter follower, or (best) an IC driver, like MIC442x or so on.
The 4981 is present in most Lambda 600 power supplies, as you can see, 600W.
The 4981 is present in most Lambda 600 power supplies, as you can see, 600W.
Yes, I think I would need a separate mosfet driver because the IC can only supply 400mA. Do you have any experience with the LT1248?
No, unfortunately no. L4981 is far one of the most viewed in industrial environments. Also, there are some rare combos (PFC + SMPS) like FAN4803, ML4824, etc, also found in newer PC SMPS's. They appear to be reliable.
In fact, I suspect that the reliability of an SMPS or PC, depends mainly in the quality and type of resistor(s) used in the source of the MOSFET for current sensing, being metal film resistors a time bomb, and in the quality of the electrolytic capacitor at the VDD of the PWM controller. Largely, most of of the faulted SMPS's and PFC's are becaus eof this such 2 components. The rest, are usually trouble free.
Then, or use good wire resistors, or a current transformer. As I said, metal film are a headache in such place.
In fact, I suspect that the reliability of an SMPS or PC, depends mainly in the quality and type of resistor(s) used in the source of the MOSFET for current sensing, being metal film resistors a time bomb, and in the quality of the electrolytic capacitor at the VDD of the PWM controller. Largely, most of of the faulted SMPS's and PFC's are becaus eof this such 2 components. The rest, are usually trouble free.
Then, or use good wire resistors, or a current transformer. As I said, metal film are a headache in such place.
I will check my Rsense I think it's a metalfilm resistor which could be adding a lot of inductance. I will buy some L4981 in case this doesn't work. Can you recommend me some cores? I got some cores with a 170mm^2 effective area and 1mm gap from some old equipment, do you think that would be enough for a 450W PFC?
Here is my PCB layout, this is bottom layer where there is the PFC IC and almost all components and routing:
And this is the top layer where I tried to keep it as a GND plane:
I marked R75 which is the current sensing resistor, you can see is very close to the source of the MOSFET and goes straight to the GND plane through a via (the resistor is in vertical position). The track that goes from the source to pin 4 is longer but I don't know if long enough to cause any issues, what do you think? Tomorrow I will check if the resistors are metalfilm (I'm 90% sure they are).
And this is the top layer where I tried to keep it as a GND plane:
I marked R75 which is the current sensing resistor, you can see is very close to the source of the MOSFET and goes straight to the GND plane through a via (the resistor is in vertical position). The track that goes from the source to pin 4 is longer but I don't know if long enough to cause any issues, what do you think? Tomorrow I will check if the resistors are metalfilm (I'm 90% sure they are).
For now, if you have external 12v, you could use that, to power the IC and don't bother with secondary and related circuit
Do you think the secondary is causing problems? Because I have a stable voltage there as far as I can see.
why do you say that? I find it very easy, not much you have to do, to make it work
if you have that voltage within specs, then probably no, I just hope you are not hitting any overvoltage protection.
for R75, what kind resistor do you have? since on that node you have high inductance, there shouldn't be that kinda waveform there, something is doing that. Pafi sure was right, that you are getting "over current"
If this will keep on going on for you, I will dig out my pfc and check waveforms. I don't think they would be similar.
Last edited:
OK, I will try powering it from an external supply just in case. It's a metalfilm resistor I'm almost sure and they seem to have enough inductance to cause problems, I will get some carbon or ceramic resistors.
Having some reference waveforms would be great!
metalfilm are actually ok. wire wound wouldn't work aka ceramic, at least on outside (5w+)
Also, you can use even higher value resistor for now, lets say 1-4.7R, 1/4w would work I guess. Since you have not powering any load, but fighting to even power is up with very light load
Explain that to me, how is this possible? you have about 230V input, that would have to give you a minimum of 300V+ since rectified main would give you that. Only thing in between input and output is bridge (which is probably not the problem) and inductor and its diode (which again is not the problem IMO)
The problem is not usualy the inductance. Their inherent low overload capability causes they to burn easily. As the resistor opens, all semiconductor stage is burned too. More over during the firts tests.
If you have 285V, then the PFC is inoperative. Post the schematic if you can, I didn't build any PFC, but I repaired tenths of them. This way, perhaps I may help you.
Luka: The main advantage of the PFC is to draw almost senoidal current from power lines, not to increase the output voltage. The boost topology is the most used because of its easily of design (sorce of the MOSFET at ground potential), low cost and simple topology, and the input from an inductor becomes easy to filter power pulses.
If you have 285V, then the PFC is inoperative. Post the schematic if you can, I didn't build any PFC, but I repaired tenths of them. This way, perhaps I may help you.
Luka: The main advantage of the PFC is to draw almost senoidal current from power lines, not to increase the output voltage. The boost topology is the most used because of its easily of design (sorce of the MOSFET at ground potential), low cost and simple topology, and the input from an inductor becomes easy to filter power pulses.
Last edited:
Luka: look at this, pse:
https://cooltechpc.com/articles/power-factor-correction-pfc
http://www.vicorpower.com/documents/application_notes/an1_active-pfc.pdf
Enjoy‼
https://cooltechpc.com/articles/power-factor-correction-pfc
http://www.vicorpower.com/documents/application_notes/an1_active-pfc.pdf
Enjoy‼
yes, but having only 3 components in power path between IN/OUT, two of them being diodes, which can be checked very easily, you are left with inductor, which is or is not correct in this circuit.
What component? inductor?
Yes and no, could be, but can't do much with what it has. Still that doesn't explain low voltage on output, since you can leave PFC turned off and still have full rectified AC voltage on output (Above that is PFC working range)
🙄 first post?
Thank you, but nothing new to me 😱
The Vcc pin has a 36V zener so I think I should be within specs. The minimum voltage should be 13V according to the datasheet (Startup Threshold).
Now I changed R75 for a 4.7Ohm 1/4W carbon resistor.
Here are some pictures of the real PCB (it's a mess):
So here are new measurementes with R75 being a 4.7 Ohm carbon resistor. This is on R75:
Pin 4:
Rectified AC:
Vout:
Should I try maybe reducing the 1mH inductance and also using some ferrite core with gap instead of the toroids I'm using now?
- Status
- Not open for further replies.