I managed to get one of my pcbs working well at 5 amps and 25 to 35 volts output. Above 35 volts the regulation starts to drop out.
I built up another but with half as many windings on the transformer to see if I could get some more current. But I noticed in the spread sheet that I don't know how to fill in "Planned induction in core" which is in Tesla. How do I work that out ?
I built up another but with half as many windings on the transformer to see if I could get some more current. But I noticed in the spread sheet that I don't know how to fill in "Planned induction in core" which is in Tesla. How do I work that out ?
The new pcb with less turns on the transformer was performing poorly with lots of ripple.
When I looked carefully I could see it was 100Hz ripple so my primary smoothing is inadequate.
When I looked carefully I could see it was 100Hz ripple so my primary smoothing is inadequate.
Whoops the last time I blew the mosfets it also took out a pcb track near the smoothing capacitors.
I used my new isolation transformer to look at the primary side and there was no smoothing on the DC line at all, its amazing it was doing anything !
I am now getting +/- 40 volts DC at 5 amps.
I used my new isolation transformer to look at the primary side and there was no smoothing on the DC line at all, its amazing it was doing anything !
I am now getting +/- 40 volts DC at 5 amps.
I did some experimentation tonight with the control circuit resistors.
I found I could get a range of voltages out if I changed Rmax to a much lower value.
I can get from 25 volts to 40 volts out by using a preset on the secondary side to the tL431.
I found I could get a range of voltages out if I changed Rmax to a much lower value.
I can get from 25 volts to 40 volts out by using a preset on the secondary side to the tL431.
Last edited:
Decreasing the Rmax increases the maximal switching frequeny. If it is set too hugh, you will loose ZVS switching, and efficiency will fail.
Nigel. The LLC converter is intented to regualte the output at a certain value. This value is in the spreadsheet. It is NOT a PWM converter which you can use easily to set any output votlage you would like.
If you have an isolation transformer check the lower FET gate and the half bridge, as I said. If you psot soem waveforms I can easily tell it is ZVS or not.
My deadtime is around 1uS so I cant see the gate signal when the frequency is very high.
Here is a picture of lower gate and half bridge signal.
Here is a picture of lower gate and half bridge signal.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
I experimented a bit more tonight with varying the output voltage.
While I can vary the voltage a little the load also dictates what the output voltage range will be.
So keeping the output voltage set is the best way to go and just use the pot as a fine trimmer.
While I can vary the voltage a little the load also dictates what the output voltage range will be.
So keeping the output voltage set is the best way to go and just use the pot as a fine trimmer.
Hi all,
Now I have a project with IRS27951, a power suply 500W -14.4 VDC output:
+ Input range: 350-420VDC.
+ Fr = 120kHz,
+ Fmin: 80kHz
+ Fmax: 160kHz
+ Fss: 200 kHz
This is my trouble when I debug my design:
+ if I dont power high voltage on Drain of highside MOSFET (Vin), everything is work well (Fmin, Fmax, Smooth wave, HO&LO is 46% duty cycle...)
+ When I power high voltage on drain of highside MOSFET - only highside and lowside MOSFET is connectted, Lr & Cr were removed (open LLC stage) - ouput of HO & LO is fail, HO duty cycle is longer than LO duty cycle (higher Vin, longer HO duty cycle)
Ex:
Vin = 0 VDC everything is good: HO:LO = 46%:46%
Vin = 50 VDC: HO:LO = 48%:44%
Vin = 200VDC: HO:LO = 50%:42%
Vin = 400VDC: HO:LO = 60%:33%
By constrast, in datasheet of IRS27951, Duty cycle of HO&LO is 50:50 (of cousre minus dead-time). What happen with my design (I tried to replace other IRS27951 but they were fail, too) ???
Can any body help me?
Please !!!
- Status
- Not open for further replies.