It seems it's not possible because Pd splits into 3 parts: Pload, P due to Vod, and P due to quiescent current.
So I do by iteration: switching nominal Vac from 20 to 21 makes Pd jump from 46 to 51W (respectively 129 & 142VA rating) for all other variables stuck: 4Ω/50mA/6dB (and Vbridgedrop=2V).
I still have to check the effects of Vrating on Vac at various PSrms .
So I do by iteration: switching nominal Vac from 20 to 21 makes Pd jump from 46 to 51W (respectively 129 & 142VA rating) for all other variables stuck: 4Ω/50mA/6dB (and Vbridgedrop=2V).
I still have to check the effects of Vrating on Vac at various PSrms .
Good reading for geeks here: Application Notes on Rectifier Transformers @ https://toroid.com/Technical-Topics/Technical-Topics
I'm no more sure of my max Pd
I'm no more sure of my max Pd
I'm no more sure of my max Pd
The math I show on my website for calculating Pd assumes symmetrical sine wave operation. What matters there is the supply voltage and the output swing. The quiescent power dissipation adds a few watt. My math will get you close to the actual result - and if anything probably overestimates the VA rating a bit.
If you want higher precision in the result, I suggest setting up a simulation. You'll find that the dissipated power and power drawn from the transformer depends on the supply voltage, DC resistance of the transformer, rectifier diode characteristics, supply capacitance, the power amp, and the signal driving the power amp, just to name a few.
The VA rating also depends, to an extent, on the cleanliness of the mains voltage. A simulation would likely use a sine wave for the mains voltage, but in reality, the mains voltage is generally a rather distorted sine wave. So you can set up a measurement if you really want the last few digits of precision.
Regardless of your method, you'll end up with a number for the required VA rating. Let's say that number is 121.23 VA per channel. I just plucked that number from thin air as an example. Now you have to pick a transformer from that. You'll likely be choosing between a 100 VA type or a 160 VA type.
The 100 VA type is too small according to your math, but how long do you plan to keep your amp running at full power. 24/7? Probably not. Maybe you redo the math using a higher crest factor than 6 dB and find you now need a 95.76 VA transformer. So 100 VA is probably OK.
160 VA would certainly be enough according to the math, but it's more expensive than the 100 VA type. Would it be worth the additional expense? Maybe not from a technical perspective, but maybe from a marketing perspective.
There's also the little detail that most amps deliver maybe 100 mW - 1 W under most listening conditions. You don't need a very big transformer to support that.
While I admire your curiosity, I think you're getting lost in the decimals here. If your goal is to build an amp, stop and reorient yourself for a bit. If your goal is to understand, I suggest setting up a simulation (hint: the LM3886 TINA-TI model does not model the supply current, so you'll either have to use a different LM3886 model or model the supply current yourself).
Tom
A bit easier way:
Transformer va subtract 10% for estimated power factor = watts out of the transformer.
That subtract estimated amplifier efficiency reversed--60% efficient amp, so then subtract 40%-- is expected output power out of the amplifier.
Also bear in mind that whatever doesn't get sent to speaker probably heats up the heatsink and enclosure. Indeed it does go somewhere, and that's where.
Please do double-check with such practical estimates!
That's because if you don't put enough power in, then your expectations of power out, will be met with gross distortion.
So, do your figures pass the most practical checkup?
Transformer va subtract 10% for estimated power factor = watts out of the transformer.
That subtract estimated amplifier efficiency reversed--60% efficient amp, so then subtract 40%-- is expected output power out of the amplifier.
Also bear in mind that whatever doesn't get sent to speaker probably heats up the heatsink and enclosure. Indeed it does go somewhere, and that's where.
Please do double-check with such practical estimates!
That's because if you don't put enough power in, then your expectations of power out, will be met with gross distortion.
So, do your figures pass the most practical checkup?
Hi all
Happy new year
Yes I want to build. But I don't want to make mistake: I don't want to be one day in the situation where I'd regret I was too lazy to understand when I decided for the transformer size. The point is that datasheet for NS/TI and Tom speak Vdc, but when it comes to building we have to deal with Vac. I really admit we don't need a single decimal of this math final result. But I'm not sure how was the transfo partial load (the so called "regulation") taken into account, knowing other power parameters are pure decisions that may be met all together: max Tamb, Crest factor and heatsink Rth being given... what about when mains supplier allows himself a continuous +10% (not even talking of the day when they decide to switch from nominal 230 to 250)... just the day I decide to use my amp as a power sinewave generator under the sun
.
It seems we have differences in transfo manufacturers: I derives this from the differences in the weight of similar builds (I mean models with shield) between e.g. ANTEK and TOROIDY, with spec sheets far less detailled than the 3886 one is.
OK guys, for now I have to spare cents for some time. I'll be back I hope near the end of february.
Bye bye
Happy new year
Yes I want to build. But I don't want to make mistake: I don't want to be one day in the situation where I'd regret I was too lazy to understand when I decided for the transformer size. The point is that datasheet for NS/TI and Tom speak Vdc, but when it comes to building we have to deal with Vac. I really admit we don't need a single decimal of this math final result. But I'm not sure how was the transfo partial load (the so called "regulation") taken into account, knowing other power parameters are pure decisions that may be met all together: max Tamb, Crest factor and heatsink Rth being given... what about when mains supplier allows himself a continuous +10% (not even talking of the day when they decide to switch from nominal 230 to 250)... just the day I decide to use my amp as a power sinewave generator under the sun
.It seems we have differences in transfo manufacturers: I derives this from the differences in the weight of similar builds (I mean models with shield) between e.g. ANTEK and TOROIDY, with spec sheets far less detailled than the 3886 one is.
OK guys, for now I have to spare cents for some time. I'll be back I hope near the end of february.
Bye bye
Transformer regulation has two effects that require a bit of planning to get to your required targets.
a.) A high regulation results in a high voltage when the amplifier is off load or when secondary fuses have blown.
This affects the voltage seen by the smoothing capacitors, They must be rated for the highest voltage they will see in operation. It also affects the amplifier' highest voltage while a very small signal is present.
Predicting the worst case PSU voltage taking account of mains supply voltage and near zero load, allows one to select adequately rated capacitors and not exceed the amp's maximum voltage
b.) A high regulation results in a high sag/droop in supply voltage when a sustained high output current is demanded by the load. This affects both the maximum output power and the sound is not so good, dyanmics are reduced.
This last is why many Builders use transformers that are apparently massively overspecified, where a smaller transformer will not overheat and will produce almost as much maximum power.
VA between one and two times the total maximum power for good domestic duty performance.
Vac so that maximum PSU voltage does not damage the amplifier.
a.) A high regulation results in a high voltage when the amplifier is off load or when secondary fuses have blown.
This affects the voltage seen by the smoothing capacitors, They must be rated for the highest voltage they will see in operation. It also affects the amplifier' highest voltage while a very small signal is present.
Predicting the worst case PSU voltage taking account of mains supply voltage and near zero load, allows one to select adequately rated capacitors and not exceed the amp's maximum voltage
b.) A high regulation results in a high sag/droop in supply voltage when a sustained high output current is demanded by the load. This affects both the maximum output power and the sound is not so good, dyanmics are reduced.
This last is why many Builders use transformers that are apparently massively overspecified, where a smaller transformer will not overheat and will produce almost as much maximum power.
VA between one and two times the total maximum power for good domestic duty performance.
Vac so that maximum PSU voltage does not damage the amplifier.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.