The board I have is the latest version that can take up to 4 LEDs. Just finish doing it, i put in 2 green LEDs (D302A and D302B) and jumper the other 2 (D303 and D304). Supply Vac=7.2, after rectification is 8.99Vdc, R301 is 20ohm, out put I tie a 2.2k ohm resister to it.
Power it up and adjust the output to 3.3V. Voltage measure across LED at D302A is 1.92v, D302B is 1.93v.
The current I stated is more than what is needed. I shunt about 70- 100mA each. The tripath amp I just follow spec, that is maximum drawn current.
Power it up and adjust the output to 3.3V. Voltage measure across LED at D302A is 1.92v, D302B is 1.93v.
The current I stated is more than what is needed. I shunt about 70- 100mA each. The tripath amp I just follow spec, that is maximum drawn current.
See this easy graphical explanation so you can estimate any constant current source setting. The 1st graph below is the IRF9610 ID current VS Vgs curve. The higher the current the more the Vgs. The two vertical lines are the voltage of 3 and 4 leds groups. Say you need to target 2 points. One is 100mA, the other is 0.96A. You need to pick horizontal lines that meet the curves to the leds lines. You see that after 0.48A the 3 leds like yours come close to the curves. So better use 3 leds for under 0.48A and 4 leds for over. Or stronger 3 leds. The horizontal lines represent the voltage across the setting resistors. One line is at 100mA and is about 1.4 Volt long. Ohm's law says 1.4V/0.1A=14 Ohm. That is R101,201,301 in the PCB. Vrset2 line is about 1.9 Volt long. 1.9V/0.96A=1.98 Ohm.
Every 9610 will not be exactly same, temperature on every sink and setting will not be the same, every different Mosfet type will have other curves to look, even IRF610 has different. But with this graphical method you will always be in the football field. Just allow enough current on top of your load demands in calculation to allow for tolerances, and to have good extra for the regs to work better. The region up to 0.35A gives the steadiest current for temperature variations. Thankfully the positive temperature tracks well up to 0.72A. On hot sinks and higher currents better draw horizontal Vrset lines stopping between those two right hand curves. So your calculated current does not end up less when the sinks come up to temperature.
Always read the Safe Operating Area when planning strong currents and voltages to know the limits of each type (2nd graph=IRF9610 SOA).
Every 9610 will not be exactly same, temperature on every sink and setting will not be the same, every different Mosfet type will have other curves to look, even IRF610 has different. But with this graphical method you will always be in the football field. Just allow enough current on top of your load demands in calculation to allow for tolerances, and to have good extra for the regs to work better. The region up to 0.35A gives the steadiest current for temperature variations. Thankfully the positive temperature tracks well up to 0.72A. On hot sinks and higher currents better draw horizontal Vrset lines stopping between those two right hand curves. So your calculated current does not end up less when the sinks come up to temperature.
Always read the Safe Operating Area when planning strong currents and voltages to know the limits of each type (2nd graph=IRF9610 SOA).
Attachments
Ar....... made a mistake while constructing my 3rd board, put BC550C at Q103 and 2SK117GR at Q104, it should be the other way around. Now no out put even Dxx2A, 2B and 3 light up. Dxx5 no light. I tried replace Q103 with new 2SK117GR and reuse back the BC550C at Q104. But same result, no output.
Anyone can tell me which components should I change after a mistake like that?
Thanks.
Anyone can tell me which components should I change after a mistake like that?
Thanks.
I have no more BC550C, need to order locally from Element14, but they don't have stock for it. Can I use this ON SEMICONDUCTOR|BC547CZL1G|TRANSISTOR, NPN, TO-92 | element14 Malaysia ?
I plan on using Salas's regs for the legato 3.1, with a supply voltage of 15V with constant load of ~ 350mA.
" You must use 2 wires from +F, +S and two wires from F0,S0 points shown on board. Or F-, S- & 0S, 0F on negative polarity. The +F, +S wires meet at +V side of a load circuit, the F0, S0 at ground side. F-, S- meet at the -V side
of a load circuit, the 0S, 0F at ground side."
So i need a + AND - regulator for a single PWR section? And if i wish to use separate PSUs for the L, R channel and bal-SE stage, then i need a total of 6 regulators (3+ 3-)? Is there any real point in using so many regulators?
"Transformer 50VA. 80VA hot-rod. Fuse to Tx primary rating. Calculate min 5VDCin more than DCout"
I don't exactly understand this.. the 350mA load = hot rod right? So i need 80VA transformer with 20v voltage rating (since i want +/- 15DC V out). Can't seem to find a 20V, will 25V do?
I'm also confused about the zener vs norton. Am i supposed to choose only one configuration? Which one offers higher performance?
Thank you
" You must use 2 wires from +F, +S and two wires from F0,S0 points shown on board. Or F-, S- & 0S, 0F on negative polarity. The +F, +S wires meet at +V side of a load circuit, the F0, S0 at ground side. F-, S- meet at the -V side
of a load circuit, the 0S, 0F at ground side."
So i need a + AND - regulator for a single PWR section? And if i wish to use separate PSUs for the L, R channel and bal-SE stage, then i need a total of 6 regulators (3+ 3-)? Is there any real point in using so many regulators?
"Transformer 50VA. 80VA hot-rod. Fuse to Tx primary rating. Calculate min 5VDCin more than DCout"
I don't exactly understand this.. the 350mA load = hot rod right? So i need 80VA transformer with 20v voltage rating (since i want +/- 15DC V out). Can't seem to find a 20V, will 25V do?
I'm also confused about the zener vs norton. Am i supposed to choose only one configuration? Which one offers higher performance?
Thank you
Not really, but some prefer an all out spending no object approach. There can be some squeeze out best VS law of diminishing returns arguments.
Use Norton Vref, more predictable for noise than what you can chance on with cheap available zeners.
I plan on using Salas's regs for the legato 3.1, with a supply voltage of 15V with constant load of ~ 350mA.
" You must use 2 wires from +F, +S and two wires from F0,S0 points shown on board. Or F-, S- & 0S, 0F on negative polarity. The +F, +S wires meet at +V side of a load circuit, the F0, S0 at ground side. F-, S- meet at the -V side
of a load circuit, the 0S, 0F at ground side."
So i need a + AND - regulator for a single PWR section? And if i wish to use separate PSUs for the L, R channel and bal-SE stage, then i need a total of 6 regulators (3+ 3-)? Is there any real point in using so many regulators?
"Transformer 50VA. 80VA hot-rod. Fuse to Tx primary rating. Calculate min 5VDCin more than DCout"
I don't exactly understand this.. the 350mA load = hot rod right? So i need 80VA transformer with 20v voltage rating (since i want +/- 15DC V out). Can't seem to find a 20V, will 25V do?
I'm also confused about the zener vs norton. Am i supposed to choose only one configuration? Which one offers higher performance?
Thank you
I'm using Legato with only two regulators: one +15V & the other -15V, only be sure install jumpers & use PWR3 input.
I have heaps of bc550c/560c, i can send you some, just sort me for postage and they are yours, will only be a couple of dollars postage for regular airmail. in fact forget postage, its not worth the trouble. i bought like 150 of each on a end of production buyout and still have a bundle. just consider it a gift and maybe one day you'll do the same for someone
indeed and its not just the CCS that will feel it, the CCS resistor will be a little toasty too
