Well, how large?
Are there any thumbrules, apart from 'The biggest that fits into the case'?
I can only get EI core transformers. No toroids. However, the ones I do get are reasonably decent, with little to no rattle and hum
If the amp is say, 200 watts, what kind of transformer am I looking at? 300 watts? 450 watts? How much is too much?
Thanks
In specifics: I want to build a pair of dual drive subs (stereo 'psuedo-subs' using regular 12" hifi woofers) using a 6 db/oct passive rolloff at about 100 Hz, and a pair of TDA 7294s running in bridge mode. The manufacturer claim is 170 W into 16 ohms in such a config, which for me is bang on target. I have a pair of Jamo E 470 floorstanders which have good bass but my room is just killing the bass. Suffice it to say that there is little I can do short of adding more low bass in the spectrum. I do NOT like using tone controls, or else that would've been the best thing to do...
Are there any thumbrules, apart from 'The biggest that fits into the case'?
I can only get EI core transformers. No toroids. However, the ones I do get are reasonably decent, with little to no rattle and hum
If the amp is say, 200 watts, what kind of transformer am I looking at? 300 watts? 450 watts? How much is too much?
Thanks
In specifics: I want to build a pair of dual drive subs (stereo 'psuedo-subs' using regular 12" hifi woofers) using a 6 db/oct passive rolloff at about 100 Hz, and a pair of TDA 7294s running in bridge mode. The manufacturer claim is 170 W into 16 ohms in such a config, which for me is bang on target. I have a pair of Jamo E 470 floorstanders which have good bass but my room is just killing the bass. Suffice it to say that there is little I can do short of adding more low bass in the spectrum. I do NOT like using tone controls, or else that would've been the best thing to do...
here the formula taught to me by an electrical engineer
first identify the amount of voltage and the current drawn by the amp and then factor in a ratio. i would show an example for better understanding
for instance i need 12 vdc unregulated and the circuit draws 1 amp
i take a 9 vac transformer that should give me about 12vdc if i used a cap input. then use 12 vdc * 1 amp to give me the va rating of the power requirements of the circuit. i should get now 12va. Of course we can't use this value the transformer would heat up and voltage would sag. So we normally factor in 1.8(really mininum). So now times 12va with 1.8. We should get a 21.6va transfromer for this circuit. 1.8 factor is a rule of thumb. Normally with experience, i tend to use a ratio of 4 instead of 1.8 as EI are not known for superb regulation.
Some transformer makers also tell us the regulation rating of the EI they make. Normally EI regulation is around 10% or more. So when full load the voltage would sag to 0.9 or 90%. Therefore with ratio of 4 i don't think u can go wrong.
first identify the amount of voltage and the current drawn by the amp and then factor in a ratio. i would show an example for better understanding
for instance i need 12 vdc unregulated and the circuit draws 1 amp
i take a 9 vac transformer that should give me about 12vdc if i used a cap input. then use 12 vdc * 1 amp to give me the va rating of the power requirements of the circuit. i should get now 12va. Of course we can't use this value the transformer would heat up and voltage would sag. So we normally factor in 1.8(really mininum). So now times 12va with 1.8. We should get a 21.6va transfromer for this circuit. 1.8 factor is a rule of thumb. Normally with experience, i tend to use a ratio of 4 instead of 1.8 as EI are not known for superb regulation.
Some transformer makers also tell us the regulation rating of the EI they make. Normally EI regulation is around 10% or more. So when full load the voltage would sag to 0.9 or 90%. Therefore with ratio of 4 i don't think u can go wrong.
Thanks guys
Maybe somewhere above 2.5 or so would be a comfortable compromise between money and power, I guess. I cannot afford 800 watt transformers to run 7294s, seems that about 400/450 watts is a good bet for 180 watts RMS, say what?
NickC... is factor of 4 really necessary? I guess for a sub it may be, but I am daunted by the 2 x 400 VA transformers required for each sub... However the regulation is also an issue to about 15% maybe (Indian construction ) so I still am thinking...
Is there anything as too large a transformer??
Maybe somewhere above 2.5 or so would be a comfortable compromise between money and power, I guess. I cannot afford 800 watt transformers to run 7294s, seems that about 400/450 watts is a good bet for 180 watts RMS, say what?
NickC... is factor of 4 really necessary? I guess for a sub it may be, but I am daunted by the 2 x 400 VA transformers required for each sub... However the regulation is also an issue to about 15% maybe (Indian construction
) so I still am thinking...Is there anything as too large a transformer??
Is 250 VA too little? Reason I ask: I have a spare 25V 250 VA lying around and it was initially planned for a nice 140 watt amp using two TDA 7294s in stereo mode, but the price of the 7294 is making me lust for the bridge amps (though the IC is one of the smallest cost items in the sub, thankfully I have the drivers). Also this particular transformer is as common as a common cold, so it is easy to get ones hands on...
Now I'm bargaining
I am succesfully running a 70 watt bass and guitar rig based on the same IC off a 120 VA 20VAC transformer.... That makes the ratio about 1.7, slightly underpowered by some standards.
But I like 300 VA. It's a nice convenient size, I just wouldn't want to waste a perfectly capable transformer.
Now I'm bargaining
I am succesfully running a 70 watt bass and guitar rig based on the same IC off a 120 VA 20VAC transformer.... That makes the ratio about 1.7, slightly underpowered by some standards.
But I like 300 VA. It's a nice convenient size, I just wouldn't want to waste a perfectly capable transformer.
I posted on this a while back and got rather disbelieving responses, but I'm running two 3886s and all associated crossover op-amps + regulators of a +/-25V 30VA transformer quite happily.
The rating was calculated for me by a bloke who designs well respected pro-audio amps for a living, so I trust him. He says it's not necessary to match the transformer to the power rating of the amp for audio - you can de-rate the transformer because of duty cycles in the signal you're running through the amps and things like that. He's turned out to be right - two 50W amps run fine off a 30VA transformer. No problems at all.
The rating was calculated for me by a bloke who designs well respected pro-audio amps for a living, so I trust him. He says it's not necessary to match the transformer to the power rating of the amp for audio - you can de-rate the transformer because of duty cycles in the signal you're running through the amps and things like that. He's turned out to be right - two 50W amps run fine off a 30VA transformer. No problems at all.
30VA is certainly something that I wouldnt recommend. For 2x50W, assuming a capacitor input power supply, if both channels peaked at the same moment, that tiny transformer would be expected to do 330VA (plus copper losses / core losses) now I think that we can safely say that it will not succeed at that. Of course, if the supply caps are nice and big (and sufficiently low ESR), this will take some of the stress off the transformer but it's still far from being a good solution. And all of this is ignoring the active x/overs & regulators etc. that are fed from that same transformer.
be approved if it was commercial. It works for "soft music" but when the party is on.... Two 50 W amps consumes 150-200 Watts and to that you must add losses in the transformer as Audiofreak says.I knew this would be contentious......
Like I said, the derating of the transformer is down to duty cycles of the program material and various other common sense factors which mean that in reality you never draw 50W continuous from each amp. This kind of specification does get approved commercially because the bloke who specified it has many many well respected pro-audio products to his name, hence why I respect the benefit of his commercial experience.
Has anybody else actually tried this lower transformer rating and had problems?
Like I said, the derating of the transformer is down to duty cycles of the program material and various other common sense factors which mean that in reality you never draw 50W continuous from each amp. This kind of specification does get approved commercially because the bloke who specified it has many many well respected pro-audio products to his name, hence why I respect the benefit of his commercial experience.
Has anybody else actually tried this lower transformer rating and had problems?
Fair point about the efficiency. On the bench supply the whole amp board draws about 100mA quiescent, so at +/-35V that's about 7W. Don't forget that this isn't an overhead though - just the minimum power ever drawn from the transformer.
Okay, I've just had a conversation with my mate who helped me out with this. The way we worked it out was this :
1. It's a biamp config, so assume the LF amp will pull the full 50W but the HF amp is very unlikely to draw more than 10W. So total power = 60W.
2. Assume efficiency is about 75%, so round up and call it about 90W maximum total power drawn.
3. Even on music with the dirtiest basslines you're unlikely to get an effective duty cycle of much more than about 30%, so de-rate accordingly and round up a bit and you get your 30VA.
The size of the smoothing caps you use will have a far greater bearing on the sonic performance of the amp than the size of the transformer (within reason obviously).
Just so this doesn't get taken the wrong way here I'd just like to point out that I'm not trying to be a smart *** or flame anyone or anything like that. I really do have the best of intentions - transformers are one of the most expensive parts of an amp so you really don't want to over-spec them especially seeing as it is going to make very little difference to the sonic performance of the amp (compared to the most other components). DIYers should really be spending money on important sonic stuff not on expensive transformers. All I'm trying to do is pass on the benefit of experience of a very good analogue design engineer in the pro-audio industry who knows about this stuff. If you want to be ultra sure, use a 50% duty cycle to de-rate once you've calculated the maximum power - this is what the absolute best pro-audio amps do (the ones that don't use switched mode supplies that is).
I hope that helps.
Okay, I've just had a conversation with my mate who helped me out with this. The way we worked it out was this :
1. It's a biamp config, so assume the LF amp will pull the full 50W but the HF amp is very unlikely to draw more than 10W. So total power = 60W.
2. Assume efficiency is about 75%, so round up and call it about 90W maximum total power drawn.
3. Even on music with the dirtiest basslines you're unlikely to get an effective duty cycle of much more than about 30%, so de-rate accordingly and round up a bit and you get your 30VA.
The size of the smoothing caps you use will have a far greater bearing on the sonic performance of the amp than the size of the transformer (within reason obviously).
Just so this doesn't get taken the wrong way here I'd just like to point out that I'm not trying to be a smart *** or flame anyone or anything like that. I really do have the best of intentions - transformers are one of the most expensive parts of an amp so you really don't want to over-spec them especially seeing as it is going to make very little difference to the sonic performance of the amp (compared to the most other components). DIYers should really be spending money on important sonic stuff not on expensive transformers. All I'm trying to do is pass on the benefit of experience of a very good analogue design engineer in the pro-audio industry who knows about this stuff. If you want to be ultra sure, use a 50% duty cycle to de-rate once you've calculated the maximum power - this is what the absolute best pro-audio amps do (the ones that don't use switched mode supplies that is).
I hope that helps.
A few things...
First, the rules that you apply I think will start getting shaky the lower you get in power. If you are talking PRO amps at several hundred watts then maybe fine. When getting closer to the lower range the rules change a little bit in favour for overrating I think. Over-head versus base consumtion et al.
Second, I think your friend is exactly what he is: A designer for commercial PRO Audio amplifiers. He is not a designer of high-end HOME gear. For the highly competative business of PRO amps I think the rule may very well be needed in order to survive. For the DIY community the rules are different. I suspect that many of the people here, just like me, think (or know by experience) that a bigger transformer brings sonic improvements. Quite a few I suspect have tested it and found that the size of the transformer indeed does not "make very little difference to the sonic performance" but rather the opposite.
Third, if using larger caps with a smaller transformer you will maybe not only load down the transformer (one of Per-Anders's favourite comments?) but also reduce the available power as the transformer will not be able to charge the caps. Severe dynamic compression. Not a good thing?
First, the rules that you apply I think will start getting shaky the lower you get in power. If you are talking PRO amps at several hundred watts then maybe fine. When getting closer to the lower range the rules change a little bit in favour for overrating I think. Over-head versus base consumtion et al.
Second, I think your friend is exactly what he is: A designer for commercial PRO Audio amplifiers. He is not a designer of high-end HOME gear. For the highly competative business of PRO amps I think the rule may very well be needed in order to survive. For the DIY community the rules are different. I suspect that many of the people here, just like me, think (or know by experience) that a bigger transformer brings sonic improvements. Quite a few I suspect have tested it and found that the size of the transformer indeed does not "make very little difference to the sonic performance" but rather the opposite.
Third, if using larger caps with a smaller transformer you will maybe not only load down the transformer (one of Per-Anders's favourite comments?) but also reduce the available power as the transformer will not be able to charge the caps. Severe dynamic compression. Not a good thing?
Hi sangram,
You asked if there's such a thing as "too big a transformer", well there may be. As the transformer's current ratings get larger, they make the filter capacitors' life worse, because of ripple current. You'd have to use more capacitors in parallel or capacitors with better ripple current rating to prevent them from getting damaged due to overheating. This usually means the screw-type "computer grade" capacitors that, in India, are either of lousy quality (local brands) or cost more than the amp itself (Sprague, etc).
First post here, hi guys.
You asked if there's such a thing as "too big a transformer", well there may be. As the transformer's current ratings get larger, they make the filter capacitors' life worse, because of ripple current. You'd have to use more capacitors in parallel or capacitors with better ripple current rating to prevent them from getting damaged due to overheating. This usually means the screw-type "computer grade" capacitors that, in India, are either of lousy quality (local brands) or cost more than the amp itself (Sprague, etc).
First post here, hi guys.
Mate, I'm only trying to help out with the benefit of someone's hard learnt practical experience.
I'd be really interested in somebody doing a blind test with using a transformer specced the same as my method, and one specced how everyone else does to see if there's any audible difference. I'd do it myself but I get the impression that the results would not be taken terribly seriously. Actually, I'll probably do it anyway so I can at least satisfy myself and say that I know first hand - don't knock it until you've tried it and all that.......
I'd be really interested in somebody doing a blind test with using a transformer specced the same as my method, and one specced how everyone else does to see if there's any audible difference. I'd do it myself but I get the impression that the results would not be taken terribly seriously. Actually, I'll probably do it anyway so I can at least satisfy myself and say that I know first hand - don't knock it until you've tried it and all that.......
Too low for high end??? Though I'm not a big fan of Crown, 0.05% THD is a pretty decent spec especially considering the large output power. If anything pro gear will tend to be overspecced because it'll get thrashed and banged about and driven into low loads and abused a lot more than most domestic or "high end" gear (or whatever it's called).
Anyway UrSv, as your sig says "Those who say it can't be done should not stop those who are doing it"
Anyway UrSv, as your sig says "Those who say it can't be done should not stop those who are doing it"
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.