Toroidal power transformer is marked at 37 volts AC per rail at 2.5 amps.
I want to power two LM3886 amplifier ICs for stereo but of course that is far too much voltage once rectified correct?
If I calculated right that would be 52vdc per rail DC before loading.
I am considering unwinding some of the secondary wire on the transformer to get to a sane voltage the chip can handle.
What would be an appropriate AC voltage on the output of the transformer before full wave bridge rectification to not go over the DC limits of the LM3886 but still provide the upper limits of audio amplification?
Also removing winding from the transformers secondary would lower voltage but raise amperage output correct?
The transformer also has a 15-0-15 rail at low amperage which will be convenient for low signal opamps.
I do not have any other toroidal transformer to work with at the moment, and this one fell on my lap so I want to make use of it.
Thanks for any ideas and help.
I want to power two LM3886 amplifier ICs for stereo but of course that is far too much voltage once rectified correct?
If I calculated right that would be 52vdc per rail DC before loading.
I am considering unwinding some of the secondary wire on the transformer to get to a sane voltage the chip can handle.
What would be an appropriate AC voltage on the output of the transformer before full wave bridge rectification to not go over the DC limits of the LM3886 but still provide the upper limits of audio amplification?
Also removing winding from the transformers secondary would lower voltage but raise amperage output correct?
The transformer also has a 15-0-15 rail at low amperage which will be convenient for low signal opamps.
I do not have any other toroidal transformer to work with at the moment, and this one fell on my lap so I want to make use of it.
Thanks for any ideas and help.
The transformer is 185 volt amps (roughly watts) adding for the other secondary, it is probably around 200 VA. This not enough for the 3886 at its full capability. If you rewound the transformer to achieve a 40 or 50w per channel amp, 200 VA would work.
Rewinding the transformer for lower voltage means heavier wire because the current must be higher to achieve the same volt amps otherwise the transformer would become hot and/or suffer voltage sag under higher load reducing the output of the amp.
Rewinding the transformer for lower voltage means heavier wire because the current must be higher to achieve the same volt amps otherwise the transformer would become hot and/or suffer voltage sag under higher load reducing the output of the amp.
Unwinding turns on the secondary does not change the amps as that is limited by the thickness of the wire.
Forget unwinding this transformer for several reasons.
1. The 15-0-15 winding is probably on top of the 37-0-37 winding so you would have to unwind all of that.
2. The 37-0-37 winding is probably single would rather than bifilar so you will have to unwind over half of it then rewind most of it back on.
3. Unwinding is more tricky than just winding new wire on, it's a potential for a tangled mess.
Just save the transformer for a discrete amp.
Forget unwinding this transformer for several reasons.
1. The 15-0-15 winding is probably on top of the 37-0-37 winding so you would have to unwind all of that.
2. The 37-0-37 winding is probably single would rather than bifilar so you will have to unwind over half of it then rewind most of it back on.
3. Unwinding is more tricky than just winding new wire on, it's a potential for a tangled mess.
Just save the transformer for a discrete amp.
you need to remove approx. 10V from 37. Data Sheet says rails need to be lower than 84V/2 at zero loads AND high line voltage.
on your toroid 37V is specified at rated current. it may be ~5% higher unloaded
NO your VA will not increase appreciably! on smaller transformers the primary windings burn out 1st.
on your toroid 37V is specified at rated current. it may be ~5% higher unloaded
NO your VA will not increase appreciably! on smaller transformers the primary windings burn out 1st.
All right just removed the tape to expose the wire and the secondary 37-0-37 is on the outside portion. The 15-0-15 is below that and then the primary at the most inner, so this can definitely be rewound.
This transformer came out of a PA amplified speaker rated at 150 or 100 watts, I forget exactly.
So if it can not provide full voltage+amperage, that is fine. I can still build a new winding to get some decent VA out of it to power LM3886 chips for something around 50 watts x2 I would suppose, considering the original use was for a 100 watt PA.
So guessing either I rewind it with some thicker wire or leave the original and trim until I get near what these chipamps can handle?
Even if the chipamps can't get their full headroom, at least I could still make a decent amplifier using this as its power supply.
This transformer came out of a PA amplified speaker rated at 150 or 100 watts, I forget exactly.
So if it can not provide full voltage+amperage, that is fine. I can still build a new winding to get some decent VA out of it to power LM3886 chips for something around 50 watts x2 I would suppose, considering the original use was for a 100 watt PA.
So guessing either I rewind it with some thicker wire or leave the original and trim until I get near what these chipamps can handle?
Even if the chipamps can't get their full headroom, at least I could still make a decent amplifier using this as its power supply.
200VA is fine for 2x60W Hi-Fi duty
200VA rating are long term heating it should support much higher peak currents with some voltage droop no biggy >I would NOT upgrade the wire*. count the turns unwinding to calculate volts per turn.
*since yer reducing secondary winding resistance the regulation will be slightly stiffer anyways
200VA rating are long term heating it should support much higher peak currents with some voltage droop no biggy >I would NOT upgrade the wire*. count the turns unwinding to calculate volts per turn.
*since yer reducing secondary winding resistance the regulation will be slightly stiffer anyways
Last edited:
I suppose if I can only get 30ish watts out of each chip that is still fine. I just want to use this transformer because it is all I have at the moment.
Rewinding is not an issue, it is well shielded between the primary coil and the outer coils and even internally shield grounded in case of failure.
Given it was used in a 100 watt RMS amplifier originally, I do not see why it can't be used for two chipamps used in stereo at 50ish watts per channel.
Rewinding is not an issue, it is well shielded between the primary coil and the outer coils and even internally shield grounded in case of failure.
Given it was used in a 100 watt RMS amplifier originally, I do not see why it can't be used for two chipamps used in stereo at 50ish watts per channel.
Thank you. This is what I was thinking. I need to do quite a bit more reading here and wait for more comments, but this was my initial thoughts.
Plan B and Update questions...
Before I even attempt to rewind the transformer I thought why not just build a regulated power supply with it to lower the voltage to what I want?
37vac (52vdc rectified/filtered) of course is far too much for each rail. Would have to lower to 35vdc or so. At that it would be 17 volts wasted in the regulator so I wonder what that computes to as heat, wasted current and so on.
Was thinking of going with the plans and schematic here which is using similar DC voltage rails into the regulators to get a sane output voltage for the chipamp..
Roman Black's gainclone amp
Since my transformers secondary is single and center tapped I figure I can use a LM317 for the positive rail and LM337 for the negative rail.
Like that circuit on that page I will beef up the current handling by using a power PNP and NPN transistor that can handle the amperage.
So the question becomes.. Is this the way to go?
Or should I just rewind the transformer?
Is there any real world benefit to a regulated supply or am I wasting effort doing so?
Before I even attempt to rewind the transformer I thought why not just build a regulated power supply with it to lower the voltage to what I want?
37vac (52vdc rectified/filtered) of course is far too much for each rail. Would have to lower to 35vdc or so. At that it would be 17 volts wasted in the regulator so I wonder what that computes to as heat, wasted current and so on.
Was thinking of going with the plans and schematic here which is using similar DC voltage rails into the regulators to get a sane output voltage for the chipamp..
Roman Black's gainclone amp
Since my transformers secondary is single and center tapped I figure I can use a LM317 for the positive rail and LM337 for the negative rail.
Like that circuit on that page I will beef up the current handling by using a power PNP and NPN transistor that can handle the amperage.
So the question becomes.. Is this the way to go?
Or should I just rewind the transformer?
Is there any real world benefit to a regulated supply or am I wasting effort doing so?
is your transformer dual windings then it might be worth it, but why not take the V reg transistors and wasted heat and apply them all to a real amplifier. consider asking about decent designs maybe a member can provide a PCB or a source for a good easy build.
like this 12/PCB http://sound.westhost.com/project3a.htm
like this 12/PCB http://sound.westhost.com/project3a.htm
Last edited:
Sorry, maybe I am tired but how did you calculate that?
If that is the numbers then I think I am still good to go using this transformer as a regulated supply, even with heat dissipation.
Assuming 37-0-37ac * 2.5 amps is 185 watts out of the transformer, divide that in two for stereo is still 92.5 watts supply power.
Okay I am still confused.
Please bare with me...
Lets say I use my 37-0-37 transformer as is (52-0-52 at DC), use both positive and negative voltage regulators to bring it down to 40 volts + and - to push rails to 80 volts + to - to reach the upper limits of the LM3886.
Transformer is 2.5 amps out.
Pushing two LM3886 chips for stereo operation on the same supply.
What kind of wattage could I expect the chips to produce in a test setting?
Sorry my brain is broken right now lol. Long night.
a regulated supply will most probably be a waste of time, money and effort. Board member AndrewT has frequently given advice on how to unwind transformers. Ask him!
Ah, alright I think I get it now.
So 2.5A no matter what is basically had from the power supply in either ideas I had (within reason).
Take voltage down to practical chipamp levels, LM3886 being +-35 volts at 8ohm for 50 watts RMS as the datasheet says.
Regardless of power supply regulation, or unwinding transformer coil, this is a fact of reality and set in stone.
So +-35 volts (70 volts) * 2.5 amps would be 175 watts, being stereo divide that by two is still 87.5 watts out of the PSU to each chip before their own losses.
So....... this can work?
Still not sure whether to use a regulated supply or not.
That seems to be a whole new area of confusion being as I read some people swear by it and others say it is a waste of time.
Hey I have the parts to build a regulated supply, and the time. This is a personal audio amplifier for myself built from scraps for fun.
So still looking for opinions on that end.
Thanks so much guys.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.