14VDC raw, if secure as lowest result for your mains lowest anticipated level, its good enough for 9VDC output with this CCS. Normally a 12VAC transformer should suffice for that. Higher voltage input linearizes better the parasitic capacitance of the Mosfet CCS provided the extra heat can be dealt with. Has to do with how many mA are set and the sinking provision. Always check that the few Watts burned on an IRF9610 won't run its core too hot because this type's junction to case heat transfer ability is bad. It has high internal thermal resistance. On the other hand its a fast very low capacitance Mosfet that brings good HF performance. A close one with 250pF instead of 175pF input capacitance but with far better thermals is Fairchild FQP3P20-ND.
I am planning to use a positive and negative board to supply 9V. I was planning to use a transformer with 15VAC secondaries. Could I run another positive board off one of the secondaries or would I have grounding issues? I am using the +/- 9V to feed a Soekris Dam-1021. I was going to use the other to feed an input board with a +8VDC regulated down to 5V and 3.3V to feed the IsolatorPi and Kali Boards and also power the isolated side of the I2S on the Dam-1021.
You can certainly steal AC from one secondary to feed another application. One wire of that secondary is already in contact with the zero buss of the +/-9V rails of course. But when the other application has a full bridge and its unipolar there are also diodes in-between them lines.
Rectification byproducts from the two apps will readily mix in the shared secondary, but those or other nuisances are supposed to be sufficiently rejected by the local regulators and bypasses. Because its a complex enough architecture regarding the number of subsystems you need to commonly power, I suggest you first try shared if you must economize for space and cost, so to see if everything will be hopefully working with no hum or harshness before some unexpectedly bad performance or hum will make you absolutely having to add a further dedicated transformer.
Rectification byproducts from the two apps will readily mix in the shared secondary, but those or other nuisances are supposed to be sufficiently rejected by the local regulators and bypasses. Because its a complex enough architecture regarding the number of subsystems you need to commonly power, I suggest you first try shared if you must economize for space and cost, so to see if everything will be hopefully working with no hum or harshness before some unexpectedly bad performance or hum will make you absolutely having to add a further dedicated transformer.
Can it be done!
Hi, I am about to build a SSLV1.1 for my Playstation 1, CD-player and for the digital part I have built the Reflector D and there was no problem doing that. It was set up for 3.6 Volt and 500 mA but now I am facing a different problem.
I need to set the SSLV up for the drive unit and need 7.6 volts, so far no problem but the current is worrying me, 0.8 to 1 amps at startup, about 250 mA idle and when playing about 450 mA.
I have used the calculator and I am putting in 1050 mA as CCS but that will leave me with some heat to trasfer when idle or in normal use. I usually leave the CD-player on for the day when listening so that it is "warmed up" all the time so long times it will stand on idle and that is what's worrying me.
In the calculator I have a Q103 that measure IDSS 3.34 mA and I was thinking of using 6xLED at 1.9 V and I will find one that measure appr. 1.75 V. That way I will have 1.5 Ohm 2W on R101 and still manage with heatsinks at min. 3.7 deg/W.
Just want to confirm my idea before I do something I am not supposed to.
Can anyone who have done this before confirm if I am on the right track, please.
Hi, I am about to build a SSLV1.1 for my Playstation 1, CD-player and for the digital part I have built the Reflector D and there was no problem doing that. It was set up for 3.6 Volt and 500 mA but now I am facing a different problem.
I need to set the SSLV up for the drive unit and need 7.6 volts, so far no problem but the current is worrying me, 0.8 to 1 amps at startup, about 250 mA idle and when playing about 450 mA.
I have used the calculator and I am putting in 1050 mA as CCS but that will leave me with some heat to trasfer when idle or in normal use. I usually leave the CD-player on for the day when listening so that it is "warmed up" all the time so long times it will stand on idle and that is what's worrying me.
In the calculator I have a Q103 that measure IDSS 3.34 mA and I was thinking of using 6xLED at 1.9 V and I will find one that measure appr. 1.75 V. That way I will have 1.5 Ohm 2W on R101 and still manage with heatsinks at min. 3.7 deg/W.
Just want to confirm my idea before I do something I am not supposed to.
Can anyone who have done this before confirm if I am on the right track, please.
Use an FQP3P20 for Q101 or an IRF9530 if handier (but slower) for better reliability at 1A or over. Especially when the sinking is limited an IRF9610 is not the first part you would trust a lot at >800mA. Has high RθjC.
Q103 is current sink to the Vref components i.e. it has to do with the shunt voltage part. Its Q102 and R108 which set the current through the three or four Leds row. Those create a steady voltage. By subtracting Q101's Vgs from it and imposing the outcome over R101, the CCS current is set as I=VR101/R101.
Q103 is current sink to the Vref components i.e. it has to do with the shunt voltage part. Its Q102 and R108 which set the current through the three or four Leds row. Those create a steady voltage. By subtracting Q101's Vgs from it and imposing the outcome over R101, the CCS current is set as I=VR101/R101.
Can not get FQP3P20 tomorrow but have IRF9530 at home so I will go for that but I wonder, IRF9610 should be able to manage 1.8 A according to datasheet and it is bust during startup that I need 800-1000 mA, when playing it consume 450mA and at idle 250 mA. I will start with 9610 and if I get ingo trouble with the sink ( witch I don't think) I can put an 9530 instead.
The 1.8A is at Tc=25C. Tc is core temperature. It has been used at 1A here before and this is its rating at Tc=100C. So the long term reliability gets compromised. There are pulse ratings and continuous ratings, there is the voltage drop across it times the current i.e. its dissipation (Watts), the thermal resistances, the sink size, the ambient temperature. So it takes full thermal analysis for precisely predicting a good limit in any given application. That not much higher than 800mA for good long term reliability with the 9610 rule of thumb I recommended is about right for many typical SSLV builds and uses.
Think I have got it all correct now but I have disco light on my pcb will all colours represented red, green and yellow! So far so good.
I ended up with a R101 at appr. 1.6 Ohm 2W so I did parallel two 3.3 Ohm 3W Panasonic so I hope I am safe powerwise the only thing that can mess up things is that those resistors ar 5% but I did not have any better resistor at home.
Now a question about caps. If I run 7.6 volt at maximum 1 A (only at start up otherwise 450 mA) why do I have to use caps rated 50-63 volt, seem overkill? Wouldn't I be safe at 16-25 volt? Thinking about it since I have 10000 uF at home for C105 but only 25V and I don't see why I can not use it otherwise I am stuck for minimum a week until I can get proper volt rating.
I am also little confused regarding the other caps since the footprint for both c103 and c104 are "covering" each other, not as the footprint for c101 and c102 where it is obvious that I can squeese in two caps after each other. Can anyone explain how the caps go, please?
By the way, I'm sitting listening to some old rock for the moment so I will introduce you all (some swedes have most likely heard them but I doubt anyone outside sweden are aware of them) to a swedish band called Wilmer X that started playing back in 1977 but they knew how to rock, check them out here (an early recording).
YouTube
I ended up with a R101 at appr. 1.6 Ohm 2W so I did parallel two 3.3 Ohm 3W Panasonic so I hope I am safe powerwise the only thing that can mess up things is that those resistors ar 5% but I did not have any better resistor at home.
Now a question about caps. If I run 7.6 volt at maximum 1 A (only at start up otherwise 450 mA) why do I have to use caps rated 50-63 volt, seem overkill? Wouldn't I be safe at 16-25 volt? Thinking about it since I have 10000 uF at home for C105 but only 25V and I don't see why I can not use it otherwise I am stuck for minimum a week until I can get proper volt rating.
I am also little confused regarding the other caps since the footprint for both c103 and c104 are "covering" each other, not as the footprint for c101 and c102 where it is obvious that I can squeese in two caps after each other. Can anyone explain how the caps go, please?
By the way, I'm sitting listening to some old rock for the moment so I will introduce you all (some swedes have most likely heard them but I doubt anyone outside sweden are aware of them) to a swedish band called Wilmer X that started playing back in 1977 but they knew how to rock, check them out here (an early recording).
YouTube
Of course you can use 25V caps at your voltage levels. The 63V caps spec is to cover the whole output range this reg can be set for. They also have less ESR when with higher voltage spec.
There are two outlines for some caps on the pcb, one round and one rectangular. The round sits inside the rectangular. Their pads are already connected in parallel. If you use C101 electrolytic you just sit it on the round outline. If you use C103 electrolytic you sit it on the round and you replace R107 with a wire link. If you go for film you sit C104 MKT/MKP on the rectangular outline and you use R107.
That band sounds alright. I didn't know them. Maybe they also do numbers across a wider pop-rock variety but this particular song had a strong rockabilly influence I think.
Do you guys mind if I ask what I need to output 30V with up to 0.6A current output capability for the Salas BIB (will be used for a SuSy Dynahi)? In the excel calculator, I set DC IN (is this supposed to be AC in?) to 37V, V out to 30V, load consumption to 300mA (because on each side of the Dynahi, there are two stages which take 300mA each on each balanced PCB, of which there are two), Constant Current to 500mA as recommended and I left everything else as normal. Is this the right way to use the calculator? Do I also need to input a value for R103 in order to make R105 require a lower resistance?
Its DCin (what your transormer's secondary gives after rectification)
Where it says current source LEDs voltage better use their Vf at 2mA because R108 degenerates Q102's IDSS to lower. If your LEDS are not random and you know their type and brand there is a Ik/Vf curve to find their Vf for various current in their datasheet. Else measure them on a jig. Some 9V battery or a bench PSU set at 9V and 20mA current limit, a 3.3K series resistor to the LED, measure Vf with the DMM probes across the LED.
Where it says current source LEDs voltage better use their Vf at 2mA because R108 degenerates Q102's IDSS to lower. If your LEDS are not random and you know their type and brand there is a Ik/Vf curve to find their Vf for various current in their datasheet. Else measure them on a jig. Some 9V battery or a bench PSU set at 9V and 20mA current limit, a 3.3K series resistor to the LED, measure Vf with the DMM probes across the LED.
P.S.
R103 is a fixed resistance portion of the total R103+VR105 to make the Vout by I to V conversion of the current sink at its bottom (Norton voltage reference). The more of that total voltage you can predict by the fixed R103 the better because a low ppm fixed resistor has superior characteristics to normal quality trimmers plus it leaves less voltage across them which translates as less mW to wear them out in the long run.
R103 is a fixed resistance portion of the total R103+VR105 to make the Vout by I to V conversion of the current sink at its bottom (Norton voltage reference). The more of that total voltage you can predict by the fixed R103 the better because a low ppm fixed resistor has superior characteristics to normal quality trimmers plus it leaves less voltage across them which translates as less mW to wear them out in the long run.
The BIB board has bridge diodes and reservoir cap included. Prefer fast and/or soft recovery TO-220 bridge diodes and select a good reputation brand's reservoir like Panasonic, Nippon Chemi-Con, Nichicon, ELNA etc. Use 28VAC secondary for 30Vout DC.
WP2773 is usually consistent
From the Reflektor-D manual:
LED types that showed proper Vf and good consistency at 2mA bias are :
Kingbright red 1.735
Kingbright yellow 1.860 (All colors come from the WP2773 line)
Kingbright green 1.921