S
SamBuca
Guest
Anyone have some quick schematics to build a bipolar 12vdc power supply? Doesn't have to move much current..just enough to satisfy 2 half watt op-amps.<P ID="signature">______________
</P>
</P>
-
Get involved.
We want your input!
Apply for Membership and join the conversations about everything related to broadcasting.
After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
If you use a disposable or false email address, your registration will be rejected.
After your membership is approved, please take a minute to tell us a little bit about yourself.
https://www.radiodiscussions.com/forums/introduce-yourself.1088/
Thanks in advance and have fun!
RadioDiscussions Administrators
> Anyone have some quick schematics to build a bipolar 12vdc
> power supply? Doesn't have to move much current..just
> enough to satisfy 2 half watt op-amps.
>
Use a transformer with abt 20~25 volt secondary into a bridge.
Connect a 1,000 uf electrolytic cap from + terminal of bridge to ground
Connect a second 1,000 uf cap from the (-) terminal of bridge to ground (the +) side of the cap going to ground)
Get two 12 volt 3 pin regulators and heatsinks (use the kind with tabs, e.g. 7812 and 7912)
The positive regulator input lead goes to the + terminal of the bridge and cap
The negative regulator input lead goes to the (-) terminal of the bridge and the ground terminal of the second cap.
Your ouput is +/- 12 vdc, as referenced to ground. For safety, use a 3-wire AC cord connected to ground. Stick everything in a project box with two LED's (one + one -, use 1 k 1/2 watt resistors to drop voltage) to show the thing is working.
On the positive regulator, the tab will be ground--mount this to a good size chunk of metal (such as a metal project box with silicon grease to dissipate the heat. Use the heat sink to dissipate heat on the other regulator, the tab will be hot.
The 7812/7912 should handle the half watt op amps if well heat sinked.
> power supply? Doesn't have to move much current..just
> enough to satisfy 2 half watt op-amps.
>
Use a transformer with abt 20~25 volt secondary into a bridge.
Connect a 1,000 uf electrolytic cap from + terminal of bridge to ground
Connect a second 1,000 uf cap from the (-) terminal of bridge to ground (the +) side of the cap going to ground)
Get two 12 volt 3 pin regulators and heatsinks (use the kind with tabs, e.g. 7812 and 7912)
The positive regulator input lead goes to the + terminal of the bridge and cap
The negative regulator input lead goes to the (-) terminal of the bridge and the ground terminal of the second cap.
Your ouput is +/- 12 vdc, as referenced to ground. For safety, use a 3-wire AC cord connected to ground. Stick everything in a project box with two LED's (one + one -, use 1 k 1/2 watt resistors to drop voltage) to show the thing is working.
On the positive regulator, the tab will be ground--mount this to a good size chunk of metal (such as a metal project box with silicon grease to dissipate the heat. Use the heat sink to dissipate heat on the other regulator, the tab will be hot.
The 7812/7912 should handle the half watt op amps if well heat sinked.
S
SamBuca
Guest
Is this what you mean:
<P ID="signature">______________
</P>
</P>
B
BobDillehay
Guest
I would use two 12 volt wall warts.
> Is this what you mean:
That's it, except for the branches with LEDs and resistors. The other end should go to ground. A couple of suggestion:
I would add ceramic caps on both sides of the regulators (one going from input lead to ground, the other going from output lead to ground). Value is not very critical, 0.1uF or 0.22uF would do. The negative regulator (79xx) is known to oscillate if these capacitors are not present. It would be wise to keep these capacitors close to regulators (IOW, with short leads).
I would also use 500Ohm resistors for the LEDs. That would give them full brightness.
You could add a couple of diodes for reverse-voltage protection if you want (you put diodes in reverse where the voltage shouldn't go, ie across input and output terminals of regulators) but it's not necessary...
Note that the pin-out for 78xx and 79xx is not the same!
Regards,
Goran Tomas
That's it, except for the branches with LEDs and resistors. The other end should go to ground. A couple of suggestion:
I would add ceramic caps on both sides of the regulators (one going from input lead to ground, the other going from output lead to ground). Value is not very critical, 0.1uF or 0.22uF would do. The negative regulator (79xx) is known to oscillate if these capacitors are not present. It would be wise to keep these capacitors close to regulators (IOW, with short leads).
I would also use 500Ohm resistors for the LEDs. That would give them full brightness.
You could add a couple of diodes for reverse-voltage protection if you want (you put diodes in reverse where the voltage shouldn't go, ie across input and output terminals of regulators) but it's not necessary...
Note that the pin-out for 78xx and 79xx is not the same!
Regards,
Goran Tomas
Yep to both.
I would also add a fuse on the AC side--even in-line type if you can't find a fuseholder.
You should be able to get most of the parts from Radio Shack, though it's getting hard to find the regulators at "the Shack."
Mouser (Mouser.com) will have everything except the project box, which you can get from Radio Shack.
I would also add a fuse on the AC side--even in-line type if you can't find a fuseholder.
You should be able to get most of the parts from Radio Shack, though it's getting hard to find the regulators at "the Shack."
Mouser (Mouser.com) will have everything except the project box, which you can get from Radio Shack.
S
SamBuca
Guest
> Mouser (Mouser.com) will have everything except the project
> box, which you can get from Radio Shack.
Hahaha...already started a new project on mouser for it
<P ID="signature">______________
</P>
> box, which you can get from Radio Shack.
Hahaha...already started a new project on mouser for it
<P ID="signature">______________</P>
> Anyone have some quick schematics to build a bipolar 12vdc
> power supply? Doesn't have to move much current..just
> enough to satisfy 2 half watt op-amps.
>
If you need the part number I'll try to find my extra unit, but I had purchased a little potted device that would take a single ended 12vdc source and turn it into +/- 12vdc. Not alot of current but it worked great. I had a utility backup FM exciter and the built-in stereo generator card needed a bipolar supply and the original supply blew up. This worked great and it's still working after four years of service.
> power supply? Doesn't have to move much current..just
> enough to satisfy 2 half watt op-amps.
>
If you need the part number I'll try to find my extra unit, but I had purchased a little potted device that would take a single ended 12vdc source and turn it into +/- 12vdc. Not alot of current but it worked great. I had a utility backup FM exciter and the built-in stereo generator card needed a bipolar supply and the original supply blew up. This worked great and it's still working after four years of service.
Note to Sam Schematics for +/-12vdc power
> Is this what you mean:
>
Sam,
The LED connections as shown will not work. The resistor/LED series circuits need to be connected between the +12 and ground and the -12 and ground.
Also, the circuit ground needs to go to the center tap of the transformer (a 25 volt center tapped transformer). The way this is drawn, you will have current distribution problems.
Neil
> Is this what you mean:
>
Sam,
The LED connections as shown will not work. The resistor/LED series circuits need to be connected between the +12 and ground and the -12 and ground.
Also, the circuit ground needs to go to the center tap of the transformer (a 25 volt center tapped transformer). The way this is drawn, you will have current distribution problems.
Neil
S
SamBuca
Guest
Revised
Here's what I gathered from the responses so far:
..which powers the line amp I found from the WMFO tech manual:
I forgot to put a fuse symbol in line with the 120vac, but I do plan to stick one in there.<P ID="signature">______________
</P>
Here's what I gathered from the responses so far:
..which powers the line amp I found from the WMFO tech manual:
I forgot to put a fuse symbol in line with the 120vac, but I do plan to stick one in there.<P ID="signature">______________
</P>
Re: Revised
> Here's what I gathered from the responses so far:
>
>
>
> ..which powers the line amp I found from the WMFO tech
> manual:
>
>
>
> I forgot to put a fuse symbol in line with the 120vac, but I
> do plan to stick one in there.
>
Sam,
Looks good to me. C4 and C6 are usually in the 1 to 10 uF range though. They are used to keep the 78xx regulators from oscillating, but what you have will probably work.
The R values in the amplifier circuit are not critical, so you can use whatever nearby values you may have in stock. If the 2.4 k resistors are changed, the two replacements need to be the same (such as 2.2 k each or 2.7 k each). The gain is the ratio of the setting of the 500 k to the 100 k so if you need more gain you can reduce the 100 k resistance. Increasing the 500 k value will likely introduce a DC offset on the output so I wouldn't do that.
The inclusion of the 330 ohm resistors is to present a balanced source impedance to the line and to prevent the op amps from oscillating due to the capacitance of the line.
I have built many line amps similar to this and they work great.
Neil
> Here's what I gathered from the responses so far:
>
>
>
> ..which powers the line amp I found from the WMFO tech
> manual:
>
>
>
> I forgot to put a fuse symbol in line with the 120vac, but I
> do plan to stick one in there.
>
Sam,
Looks good to me. C4 and C6 are usually in the 1 to 10 uF range though. They are used to keep the 78xx regulators from oscillating, but what you have will probably work.
The R values in the amplifier circuit are not critical, so you can use whatever nearby values you may have in stock. If the 2.4 k resistors are changed, the two replacements need to be the same (such as 2.2 k each or 2.7 k each). The gain is the ratio of the setting of the 500 k to the 100 k so if you need more gain you can reduce the 100 k resistance. Increasing the 500 k value will likely introduce a DC offset on the output so I wouldn't do that.
The inclusion of the 330 ohm resistors is to present a balanced source impedance to the line and to prevent the op amps from oscillating due to the capacitance of the line.
I have built many line amps similar to this and they work great.
Neil
Re: Revised
> so if you need more gain you can reduce the 100 k
Just don't go too low (<1k) as it lowers the input impedance of the circuit.
> The inclusion of the 330 ohm resistors is to present a
> balanced source impedance to the line and to prevent the op
> amps from oscillating due to the capacitance of the line.
And to limit current and protect op-amp in the event of short-circuiting the output to ground (very common).
Regards,
Goran Tomas
> so if you need more gain you can reduce the 100 k
Just don't go too low (<1k) as it lowers the input impedance of the circuit.
> The inclusion of the 330 ohm resistors is to present a
> balanced source impedance to the line and to prevent the op
> amps from oscillating due to the capacitance of the line.
And to limit current and protect op-amp in the event of short-circuiting the output to ground (very common).
Regards,
Goran Tomas
Re: Revised goran's post
> > so if you need more gain you can reduce the 100 k
>
> Just don't go too low ( The inclusion of the 330 ohm
> resistors is to present a
> > balanced source impedance to the line and to prevent the
> op
> > amps from oscillating due to the capacitance of the line.
>
> And to limit current and protect op-amp in the event of
> short-circuiting the output to ground (very common).
>
> Regards,
> Goran Tomas
>
Hi Goran,
Good comments. I would add that I never lost a line drive due to shorts, but I was using MC1458 IC's which are output current limiting protected.
I did lose them due to lightning. Mine were connected to leased direct metallic phone lines which ran 4 miles end to end. There is a simple protection scheme for this which I can post if there is interest.
Neil
> > so if you need more gain you can reduce the 100 k
>
> Just don't go too low ( The inclusion of the 330 ohm
> resistors is to present a
> > balanced source impedance to the line and to prevent the
> op
> > amps from oscillating due to the capacitance of the line.
>
> And to limit current and protect op-amp in the event of
> short-circuiting the output to ground (very common).
>
> Regards,
> Goran Tomas
>
Hi Goran,
Good comments. I would add that I never lost a line drive due to shorts, but I was using MC1458 IC's which are output current limiting protected.
I did lose them due to lightning. Mine were connected to leased direct metallic phone lines which ran 4 miles end to end. There is a simple protection scheme for this which I can post if there is interest.
Neil
S
SamBuca
Guest
Re: Revised goran's post
> I did lose them due to lightning. Mine were connected to
> leased direct metallic phone lines which ran 4 miles end to
> end. There is a simple protection scheme for this which I
> can post if there is interest.
Couldn't you just put a ferrite core on the line for transients and then shunt it?<P ID="signature">______________
</P>
> I did lose them due to lightning. Mine were connected to
> leased direct metallic phone lines which ran 4 miles end to
> end. There is a simple protection scheme for this which I
> can post if there is interest.
Couldn't you just put a ferrite core on the line for transients and then shunt it?<P ID="signature">______________
</P>
Re: Revised goran's post
> > I did lose them due to lightning. Mine were connected to
> > leased direct metallic phone lines which ran 4 miles end
> to
> > end. There is a simple protection scheme for this which I
>
> > can post if there is interest.
>
> Couldn't you just put a ferrite core on the line for
> transients and then shunt it?
>
Sam,
No, there is too much energy coming in. The ferrites will supress the higher frequencies of the strike but the low frequencies will get through. The key concept is to give the strike energy somewhere to go other than through your electronics. I think you know this since you mentioned shunting. The diodes described in the next paragraph do this.
There are two solutions I have used. The simplest is to connect diodes from each line to the power supplies. The diodes are connected to be reverse biased and only conduct if the line voltage exceeds the power supply voltage by .7 volts. This dumps the energy in to the supplies. I only use this for short runs.
The best solution I have used involved zener diodes and series resistance. This is quite involved and I will describe this if you want me to. It really depends on how long your lines are. Within a building the diodes work well. From building to building the zener design is needed.
Some will mention MOV suppressors, and they can be applied here but it must be done with care.
Neil
> > I did lose them due to lightning. Mine were connected to
> > leased direct metallic phone lines which ran 4 miles end
> to
> > end. There is a simple protection scheme for this which I
>
> > can post if there is interest.
>
> Couldn't you just put a ferrite core on the line for
> transients and then shunt it?
>
Sam,
No, there is too much energy coming in. The ferrites will supress the higher frequencies of the strike but the low frequencies will get through. The key concept is to give the strike energy somewhere to go other than through your electronics. I think you know this since you mentioned shunting. The diodes described in the next paragraph do this.
There are two solutions I have used. The simplest is to connect diodes from each line to the power supplies. The diodes are connected to be reverse biased and only conduct if the line voltage exceeds the power supply voltage by .7 volts. This dumps the energy in to the supplies. I only use this for short runs.
The best solution I have used involved zener diodes and series resistance. This is quite involved and I will describe this if you want me to. It really depends on how long your lines are. Within a building the diodes work well. From building to building the zener design is needed.
Some will mention MOV suppressors, and they can be applied here but it must be done with care.
Neil
S
SamBuca
Guest
Re: Revised goran's post
> The best solution I have used involved zener diodes and
> series resistance. This is quite involved and I will
> describe this if you want me to. It really depends on how
> long your lines are. Within a building the diodes work well.
> From building to building the zener design is needed.
My particular application for this is just rack equipment. Our 15khz telco loop has a Tellabs amp on the other end, which is causing me quite a headache right now actually. SOMETHING is introducing noise and I can't find it.<P ID="signature">______________
</P>
> The best solution I have used involved zener diodes and
> series resistance. This is quite involved and I will
> describe this if you want me to. It really depends on how
> long your lines are. Within a building the diodes work well.
> From building to building the zener design is needed.
My particular application for this is just rack equipment. Our 15khz telco loop has a Tellabs amp on the other end, which is causing me quite a headache right now actually. SOMETHING is introducing noise and I can't find it.<P ID="signature">______________
</P>
Re: Revised goran's post
>
> My particular application for this is just rack equipment.
> Our 15khz telco loop has a Tellabs amp on the other end,
> which is causing me quite a headache right now actually.
> SOMETHING is introducing noise and I can't find it.
>
Sam,
Check the signal on both ends. Disconnect the source amp from the line and check its output. It must be symmetric about signal ground meaning one volt up on wire A and one volt down on wire B. Sometimes the amp will have one output fail (call it the push) and the pull will still work. You get a signal through but it is not true differential. Noise, instead of being subtracted at the receiver will get through.
Other than this, all I can offer is that there is something wrong with the line balance on one end or the other. I had terrible problems with noise and they all were due to line balance problems such as I mentioned. Blown amps on either end or unintentional grounding of one of the line wires. One I remember is water from a dishwasher leaked to the floor below and got into the terminal block. It really messed things up.
Hope this helps.
Neil
>
> My particular application for this is just rack equipment.
> Our 15khz telco loop has a Tellabs amp on the other end,
> which is causing me quite a headache right now actually.
> SOMETHING is introducing noise and I can't find it.
>
Sam,
Check the signal on both ends. Disconnect the source amp from the line and check its output. It must be symmetric about signal ground meaning one volt up on wire A and one volt down on wire B. Sometimes the amp will have one output fail (call it the push) and the pull will still work. You get a signal through but it is not true differential. Noise, instead of being subtracted at the receiver will get through.
Other than this, all I can offer is that there is something wrong with the line balance on one end or the other. I had terrible problems with noise and they all were due to line balance problems such as I mentioned. Blown amps on either end or unintentional grounding of one of the line wires. One I remember is water from a dishwasher leaked to the floor below and got into the terminal block. It really messed things up.
Hope this helps.
Neil
S
SamBuca
Guest
Re: Revised goran's post
> Other than this, all I can offer is that there is something
> wrong with the line balance on one end or the other. I had
> terrible problems with noise and they all were due to line
> balance problems such as I mentioned. Blown amps on either
> end or unintentional grounding of one of the line wires.
> One I remember is water from a dishwasher leaked to the
> floor below and got into the terminal block. It really
> messed things up.
I had the telephone company verify their end of the operation was working correctly...everything was equally balanced. I take the Compellor set at +4 directly into the telco line...at that point the signal is clean. At the other end it comes off the demarc into the Tellabs amp. When the amp is unplugged, I get silence. When the amp is plugged in with no audio connected, I get the hum but it isn't bad. When I connect the audio lines, the hum gets worse. Classic symptoms of ground loop....but I've lifted/grounded everything. I've moved ground lines. I've switched amp cards. I've switched to entirely different amp units. I can't seem to null it out.
There's just something I'm not seeing...next trip up there I'll spend some time ripping stuff apart.<P ID="signature">______________
</P>
> Other than this, all I can offer is that there is something
> wrong with the line balance on one end or the other. I had
> terrible problems with noise and they all were due to line
> balance problems such as I mentioned. Blown amps on either
> end or unintentional grounding of one of the line wires.
> One I remember is water from a dishwasher leaked to the
> floor below and got into the terminal block. It really
> messed things up.
I had the telephone company verify their end of the operation was working correctly...everything was equally balanced. I take the Compellor set at +4 directly into the telco line...at that point the signal is clean. At the other end it comes off the demarc into the Tellabs amp. When the amp is unplugged, I get silence. When the amp is plugged in with no audio connected, I get the hum but it isn't bad. When I connect the audio lines, the hum gets worse. Classic symptoms of ground loop....but I've lifted/grounded everything. I've moved ground lines. I've switched amp cards. I've switched to entirely different amp units. I can't seem to null it out.
There's just something I'm not seeing...next trip up there I'll spend some time ripping stuff apart.<P ID="signature">______________
</P>
Re: Revised goran's post
Voltages can be induced into telco pairs from other pairs in the same cable.
Not quite sure what you can do about it. Maybe a large cap in series with each lead on the telco side of the Tellabs input.
Just guessing.
One thing confusing me, you say when the audio lines are plugged into the "audio lines the hum gets louder. Do you mean the incoming lines from tellco? What does the output circuit of the Compellor look like--is it transformerless? May be a good quality transformer at that point?
Voltages can be induced into telco pairs from other pairs in the same cable.
Not quite sure what you can do about it. Maybe a large cap in series with each lead on the telco side of the Tellabs input.
Just guessing.
One thing confusing me, you say when the audio lines are plugged into the "audio lines the hum gets louder. Do you mean the incoming lines from tellco? What does the output circuit of the Compellor look like--is it transformerless? May be a good quality transformer at that point?
S
SamBuca
Guest
Re: Revised goran's post
Well, I just rewired it again a different way. Appears the noise is coming from the amp itself. The gain on the amp has to be set to 30-40, which pulls in all the extra noise from the telco line plus the already noisy amp.
I've switched from 30vac to filtered 48vdc, I actually spent an hour trying EVERY combination of grounding (again) just incase I missed one when looking it over, and the telco verified proper voltage/balance on both ends of the line.
There's also pulsing on the line. Every so often you'll hear a mid-frequency fast pulse...similar to a telephone ringing. That tells me it's picking up the AC voltage from an adjacent line ringing in the pipe. Telco denies it.
Compellor output is transformerless differential servo balanced +20dBu at 65 ohms for a 600 ohm load. There's a homemade transformer in the wiring closet but all it has is the pinouts drawn on the side...no idea what it is.
I'm half tempted to build a a directional wifi antenna, boost the output on this 802.11b ap I have, and do a microwave STL in the unlicensed spectrum...it's only a 3-4 mile clear shot. At least we'd have some ghetto internet access at the site<P ID="signature">______________
</P>
Well, I just rewired it again a different way. Appears the noise is coming from the amp itself. The gain on the amp has to be set to 30-40, which pulls in all the extra noise from the telco line plus the already noisy amp.
I've switched from 30vac to filtered 48vdc, I actually spent an hour trying EVERY combination of grounding (again) just incase I missed one when looking it over, and the telco verified proper voltage/balance on both ends of the line.
There's also pulsing on the line. Every so often you'll hear a mid-frequency fast pulse...similar to a telephone ringing. That tells me it's picking up the AC voltage from an adjacent line ringing in the pipe. Telco denies it.
Compellor output is transformerless differential servo balanced +20dBu at 65 ohms for a 600 ohm load. There's a homemade transformer in the wiring closet but all it has is the pinouts drawn on the side...no idea what it is.
I'm half tempted to build a a directional wifi antenna, boost the output on this 802.11b ap I have, and do a microwave STL in the unlicensed spectrum...it's only a 3-4 mile clear shot. At least we'd have some ghetto internet access at the site
<P ID="signature">______________</P>
- Status