I am building an amplifier that will have voltage rails between +/-12 and +/-40v, using OPA541. I want to use fans to cool the heatsinks.
Te two fans i have are 12v, 0.29A. I want to load both rails equally, so i will need to regulate 24-80v down to 12-24v.
I thought of the two most obvious solutions: using a regulator, or a zener diode. The regulators I have found have an input rating of 40v max. I think it will be to much voltage to drop with a zener diode, and I have not used them much, so I dont know the best way to use them.
My question: how should i regulate the voltage? Should I use a ~33v zener diode (the highest rating i can get), and then a regulator (15v)?
Thanks in advance,
Matthew
Te two fans i have are 12v, 0.29A. I want to load both rails equally, so i will need to regulate 24-80v down to 12-24v.
I thought of the two most obvious solutions: using a regulator, or a zener diode. The regulators I have found have an input rating of 40v max. I think it will be to much voltage to drop with a zener diode, and I have not used them much, so I dont know the best way to use them.
My question: how should i regulate the voltage? Should I use a ~33v zener diode (the highest rating i can get), and then a regulator (15v)?
Thanks in advance,
Matthew
You will have two rails which means one regulator on each will have maximum input 40 V. You can connect the two fans in series to have a 24 V fan if they are identical. You can take any standard regulator and apply the Zener diode trick on adjust to allow higher input voltage than 40 V. Check the datasheet for e.g. LM317. Use a couple of regulators to keep dissipation down. A Zener diode regulation is most likely not very good with 0.3 A and drop of 28 to 78 V on the resistor. Probably you don't want to connect an SMPS supply on the same rails that your amp run off.
Most important question though:
Why would you need fans? Are the sinks that small?
Most important question though:
Why would you need fans? Are the sinks that small?
If you for low voltage rails like +-18V go for a regulator but anything higher you´ll dissipate quite a lot if you intend supplying the full 12V to the fans (which you probably don´t need anyway).
Supplying the fans with a little 9V wallwart can be a solution likewise.
Depending on your heatsink size you could install a switch to turn the fans off for normal listening.
An SMPS might be quite noisy but I haven´t tried myself and the OPA chips might have enough PSRR to make it inaudible.
Supplying the fans with a little 9V wallwart can be a solution likewise.
Depending on your heatsink size you could install a switch to turn the fans off for normal listening.
An SMPS might be quite noisy but I haven´t tried myself and the OPA chips might have enough PSRR to make it inaudible.
The simple answer...
Hi,
Fans aren't that critical about their operating voltage. One way to regulate your rail (repeat this for each rail) is to make a voltage reference with a 12V zener diode and a simple resistor to send a few mA through the zener (or maybe even a constant current source, if your rail voltage can vary a lot). This gives you a reference point at about 12V rel. to GND.
On the positive rail, connect this reference point to the base of a NPN driver (power) transistor (a transistor with a reasonable current gain, and capable of handling the power). The collector goes to positive rail, and the emmitter goes to the fan.
This should give you approx. 11V across the fan motor.
How about it?
Jennice
Hi,
Fans aren't that critical about their operating voltage. One way to regulate your rail (repeat this for each rail) is to make a voltage reference with a 12V zener diode and a simple resistor to send a few mA through the zener (or maybe even a constant current source, if your rail voltage can vary a lot). This gives you a reference point at about 12V rel. to GND.
On the positive rail, connect this reference point to the base of a NPN driver (power) transistor (a transistor with a reasonable current gain, and capable of handling the power). The collector goes to positive rail, and the emmitter goes to the fan.
This should give you approx. 11V across the fan motor.
How about it?
Jennice
Put a zener diode in series with a regulator to bring the input voltage down to within tolerance for the regulator.
Just make sure that you are always drawing enough current to reach the avalance point of the zener.
-Bruce
Just make sure that you are always drawing enough current to reach the avalance point of the zener.
-Bruce
Zener power
FLZapped:
I think one of the concerns is that the current through the zener, along with the woltage drop will cause too much heat and destroy the zener, if it is active in the "supply line" to the fans.
Jennice
FLZapped:
I think one of the concerns is that the current through the zener, along with the woltage drop will cause too much heat and destroy the zener, if it is active in the "supply line" to the fans.
Jennice
i was thinking of connecting the fans in serial, and using abou 15v to run the fans at about 7.5v each.
i probably dont NEED the fans, but i want to be extra safe.
i think it would be better to use a zener, as i can run them from any input voltage, but i dont know the best way to use them. do i need a series resistor before the zener?
if it is too difficult to regulate the rail voltage, i could always use a sperate (~12v) transformer to run them.
what do you think i should do?
i probably dont NEED the fans, but i want to be extra safe.
i think it would be better to use a zener, as i can run them from any input voltage, but i dont know the best way to use them. do i need a series resistor before the zener?
if it is too difficult to regulate the rail voltage, i could always use a sperate (~12v) transformer to run them.
what do you think i should do?
once the motor is moving (0.29a), the power dissipation on the regulator isn't that big. For example, if you are dropping +/- 36v down to +/- 12v, that's 24v*2*0.29=15w, give or take a few. Not too bad.
The problem is that motors usually need a much larger current to get started. 2x - 3x of current in the first second or so isn't that rare. So you are talking about 30w - 45w of power dissipation. That's a lot.
In my case, I used two fans, each 12v and 0.15amp, off a LM simple switcher. Unshielded and next to a JLH1969 (poor psrr). No noise whatsoever.
The problem is that motors usually need a much larger current to get started. 2x - 3x of current in the first second or so isn't that rare. So you are talking about 30w - 45w of power dissipation. That's a lot.
In my case, I used two fans, each 12v and 0.15amp, off a LM simple switcher. Unshielded and next to a JLH1969 (poor psrr). No noise whatsoever.
The switching frequency of NS Simple Switchers is usually between 50 kHz and 200 kHz (LM2936). The PSRR of most amplifiers at these frequencies will be quite poor. However, that shouldn't matter, because you won't be able to hear it anyway. If you really want to kill the noise from a switcher, use an inductor in series with the input. Of course, put reservoir and bypass capacitors after the inductor.
A cheaper solution in most cases would be to just add a small (4-8 VA) transformer and use a bridge rectifier to power the fans. In Bangalore, at least, transformers of that size tend to be cheaper than a switcher.
A cheaper solution in most cases would be to just add a small (4-8 VA) transformer and use a bridge rectifier to power the fans. In Bangalore, at least, transformers of that size tend to be cheaper than a switcher.
Hello
I have made a high-voltage regulator for powering some op-amp circuitry off my +/-60 volt power amp supply, it's basically the same as what Jennice has described with the transistor and zener reference. If you want I can supply you with the PCB layout, or if I manage to get my etching tank sorted out this weekend I have a few different boards to knock out and I can prolly do another one of these.
Let me know by email to be sure I catch you! Use the email thingy you get by clicking my name, or the email button below.
I have made a high-voltage regulator for powering some op-amp circuitry off my +/-60 volt power amp supply, it's basically the same as what Jennice has described with the transistor and zener reference. If you want I can supply you with the PCB layout, or if I manage to get my etching tank sorted out this weekend I have a few different boards to knock out and I can prolly do another one of these.
Let me know by email to be sure I catch you! Use the email thingy you get by clicking my name, or the email button below.
True, but I like things to be neat and I find it easier and safer (as parts can't move about) to use a PCB 🙂
If you use the ESP project, I would recommend using a Darlington transistor in place of the single device. That way both the zener and resistor can be low-power, i.e. normal stock, items.
If you use the ESP project, I would recommend using a Darlington transistor in place of the single device. That way both the zener and resistor can be low-power, i.e. normal stock, items.
if you even go so far as to look at the LM317 regulator data sheet, you'll see that it can be made to act as a "buck" regulator -- of course there are better chips to utilize for this application, but in a pinch use the LM317 to regulate down - and you won't burn watts as with a zener or other method.
ok, thanks for the ideas.
i dont want to build a switcher.
i dont really want to use a regulator.
isnt the ESP project a bit overkill? will it produce less or more heat than a zener diode on its own?
[edit]: i am going to buy a few zeners and resistors at the weekend and have a play, to see how much heat this is going to produce...
i dont want to build a switcher.
i dont really want to use a regulator.
isnt the ESP project a bit overkill? will it produce less or more heat than a zener diode on its own?
[edit]: i am going to buy a few zeners and resistors at the weekend and have a play, to see how much heat this is going to produce...
MatttCattt,
Let me know how your experiments work out.
If you find the ESP project to be overkill, you may want to settle for the small, seperate transformer. It can hardly get simpler.
I think the general concern has been the power dissipation in the zener if you send enough current through it to actually power the fans directly.
Regarding the ESP design, I would probably want to try for a higher bias resistor (less bias current through the zener), and a darlington transistor for a high current gain.
Jennice
Let me know how your experiments work out.
If you find the ESP project to be overkill, you may want to settle for the small, seperate transformer. It can hardly get simpler.
I think the general concern has been the power dissipation in the zener if you send enough current through it to actually power the fans directly.
Regarding the ESP design, I would probably want to try for a higher bias resistor (less bias current through the zener), and a darlington transistor for a high current gain.
Jennice
It will produce exactly the same amount of heat! It's still dropping the same voltage and so the current must be the same. Power equals voltage x current 🙂
You will have to use high-power zeners and resistors if you want to do it using just a resistor and zener.
Re: Zener power
A 5v 5w zener will work just fine, remember, I said in series with another regulator. The zener is just to bring the input voltage to within the safe specs for the regulator.
Jennice said:
A 5v 5w zener will work just fine, remember, I said in series with another regulator. The zener is just to bring the input voltage to within the safe specs for the regulator.
Good point, FLZapped...
My brain cell was still spinning around itself to create it's own simple way of lowering voltage to supply the fans directly.
anyone got a handful of brain cells to spare...? I think I scattered mine around the living room floor when playing Destruction Derby on the PS2 console over the weekend. 😀
(And my g/f won over me
)
Jennice
My brain cell was still spinning around itself to create it's own simple way of lowering voltage to supply the fans directly.
anyone got a handful of brain cells to spare...? I think I scattered mine around the living room floor when playing Destruction Derby on the PS2 console over the weekend. 😀
(And my g/f won over me
)Jennice
- Status
- Not open for further replies.