I'm running into the exact same issue, with the circuit configured the same way as on the application notes on the data sheet (ref 5V input to 300V ou tput). The gate signal is on for a very short duration. Duty cycle is aroun d 1%. Anybody found a fix to this issue?
PS: I got this circuit to work a few days back, and it mysteriously stopped working today.
Do you have a source current sense resistor? A little inductance there can mess things up. Data sheet fig 5.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
the
his
l
lf
ame way as on the application notes on the data sheet (ref 5V input to 300V output). The gate signal is on for a very short duration. Duty cycle is ar ound 1%. Anybody found a fix to this issue?
ped working today.
No, I do not have a source current sense resistor.
That's not the same as Tim's problem.
Assuming you mean the circuit on pg 30 here
Short-cycling suggests the current-limit is getting tripped early. - Missing 22r-10nF filter on the sense input? - Shorted transformer? - Windings reversed? - open sense resistor?
Cheers, James Arthur
The chip architecture assumes one.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
the
this
nal
elf
same way as on the application notes on the data sheet (ref 5V input to 300 V output).
%. Anybody found a fix to this issue?
pped working today.
Yes, Too figured that the current limit is getting tripped early. I have bo th the 22r and 10nf cap.Soldering connections for the sense resistor are al so good (it's not open). I'm running a relatively new transformer (coilcraf t da2032) and just checked the connections on it. Looks ok.
be the
e this
hen
eral
tself
e same way as on the application notes on the data sheet (ref 5V input to 3
00V output). 1%. Anybody found a fix to this issue?topped working today.
both the 22r and 10nf cap.Soldering connections for the sense resistor are also good (it's not open). I'm running a relatively new transformer (coilcr aft da2032) and just checked the connections on it. Looks ok.
Well then it's time to dive in with a 'scope and see where it's going wrong.
You should see a sharp up-then-down Cgs glitch across the sense resistor at turn-on time, followed by a nice slow ramp. Do you?
Cheers, James Arthur
e:
:y be the
use this
then
everal
itself
the same way as on the application notes on the data sheet (ref 5V input to 300V output).
nd 1%. Anybody found a fix to this issue?
stopped working today.
e both the 22r and 10nf cap.Soldering connections for the sense resistor ar e also good (it's not open). I'm running a relatively new transformer (coil craft da2032) and just checked the connections on it. Looks ok.
NopeI dont, I see a spike about 600mV pk-pk, but it doesn't look anything l ike the waveform that's added for reference on the data sheet. The weird th ing is this circuit was left untouched for a week or so after it worked las t, so it's very surprising that it isn't working. Also, could parasitics be an issue? I'm using two seperate protoboards (one for the power stage and one for the controller, bad idea, yes, but I needed to look at both circuit s in isolation) with some fairly long jumper cables to make connections bet ween the boards.
LT suggests milliohms of Rsense. It won't take many nH of stray inductance, or resistive ground loops, to make trouble. This is not the sort of circuit that's suited to protoboards, if indeed any are.
Separating the boards will make things much worse.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
:
ote:
te:
may be the
I use this
l, then
several
er itself
d the same way as on the application notes on the data sheet (ref 5V input to 300V output).
ound 1%. Anybody found a fix to this issue?
ly stopped working today.
ave both the 22r and 10nf cap.Soldering connections for the sense resistor are also good (it's not open). I'm running a relatively new transformer (co ilcraft da2032) and just checked the connections on it. Looks ok.
rlike the waveform that's added for reference on the data sheet. The weird thing is this circuit was left untouched for a week or so after it worked l ast, so it's very surprising that it isn't working. Also, could parasitics be an issue?
he controller, bad idea, yes, but I needed to look at both circuits in isol ation) with some fairly long jumper cables to make connections between the boards.
That simply isn't going to work. Solder the two boards' ground planes (they do have ground planes, right?) directly together and you might have a chance.
Cheers, James Arthur
Of course a solid ground plane is basic here.
The LTC part suggests absurdly small current-sense resistor values, which can make ground plane drops a problem. One trick is to use a larger current-sense resistor, and a voltage divider right at the regulator chip. That adds just one part to the RC spike filter, and greatly reduces ground loop problems. Assuming the sense resistor dissipation stays reasonable.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.