A stupid power supply question

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi all,

I have two 2x30v 225VA toroid transformer and I want to use them for a gainclone power supply.
The question is;
May I use them like this;
An externally hosted image should be here but it was not working when we last tested it.

Of course I will use two of them. However I am not sure that it will give result as conventional bridge rectifiers.. I used to make like this power supplies in high school but it was a long time ago. And now I am not sure that I can suck 3,75A per rails (225/60=3,75) with this configuration..
Thanks in advance for replies..
 
TheMG said:
Why would you even want to use anything other than a full-wave bridge rectifier for a standard linear power supply?
One reason (that does *not* apply here, though) is limited current capabilty of the xformer due to used wire diameter for the secondaries. With old xformers for tube stuff you sometimes cannot load them with a full bridge.

- Klaus
 
Don't connect the two xformers as shown: each one will supply only direct current, and you cannot source more than about half the rating, or about 1.5 A.
Connect the two secondaries in series, and use a full wave bridge rectifier.
pikkujöpö:
this indeed is a full wave rectifier, but it isn't a bridge.
That configuration works well with center tapped xformers, and gives close to 0.9 the rating of a bridge.
 
No..
This question is really STUPID.
Because as mentioned, in high school years, I made a lot of this type full wave rectifiers without a problem.
But since DIY audio began in my life, I always use bridge type rectifiers and I forgot to calculate current capability of that type rectifiers. Means I was not sure if it can source 3,75A or 7,5A with this conf...
So if you forgot something really basic and already known then its STUPID...
:D
Thanks
 
In fact, my real question is some more "clever"...
I have two of 225VA, 2 x 30v (independent secondaries) toroids. And I want to use them for my regulated gainclone project.
In the begining the project was like that;
An externally hosted image should be here but it was not working when we last tested it.

As you can understand, its for one dual secondary (or two single secondary) transformer. It will feed two mono LM3875 gainclone modules. Each will sink 3,14A RMS and 4,43A peak at full volume. So if RMS values will be taken then total transformer need will be ~375VA... However I have two 225VAs..
And my "IDEA" started at this point. If I have two 225VA tranformers that means I have 450VA transformer which is fairly enough for my project. The problem is; I need two 30v secondaries that can source 6,5 amperes min. However I have four 30v secondaries each can source 3,.75A...
So this solution has born. Less component more power...
Do you think it works?
 
Hi,
parallel the outputs of each toroid.
That gives you 30Vac & 7.5Aac.
Now feed each of these to a bridge rectifier.
Regulate the 44Vdc down to a suitable voltage for your pair of chipamps.
What average current draw do you expect?
What average voltage drop do you expect?
Have you calculated the heatsink and operating temperatures of the regulators?
What if the mains runs at near maximum tolerances for more than a few minutes?

But I see a problem.
You state 4.4Apk into a load. This seems to me to be a resistive load. Your speakers are reactive. The transient current from 35Vpk into an 8ohm speaker can approach 12Apk and into a 6ohm speaker nearly 16Apk.
Where is this peak current to come from? C3 & C4? and certainly not from the regulators.

Can you explain why you think you need 375VA?
If you do need 375VA then what ripple do you expect across the smoothing caps, C1 & 2?
What is the drop out voltage of the regulators?
What is the minimum mains supply voltage?
Have you calculated if the 30Vac can give the required minimum voltage at the regulator inputs for worst case conditions?
 
Right,

16A comes from nowhere..
As you can remember we already discussed that matter before. However if I take peak values and add something for inductive needs then I will need 60x16=960VA for just ONE CHANNEL!

So I am going to label my amplifier "60W RMS into 6R load" as usual.

I hope I will never need real 60W for my 6R speakers...

PS: Output voltage will be 33+33v after regulators. Then as my calculations; output power will be 62,79W RMS into 6R load. Under that conditions, a resistive 6 ohm load will sink 4,58A peak and 3,25A RMS from the supply. I know my "poor" supply is not enough even for peak currents. However I am sure (hope) my toroids will source a bit more than 3,75A..
So I am not worrying about the inductive peak current values... Does anyone worry?
 
No?
My calculations not correct? You say 16A will be sinked for 6 ohms, my speakers are 6 ohms, so I will need 16A peak, and my transformer is 30+30=60V that makes 960VA, right?

and the "nine" answer;

What average current draw do you expect?
- My new speakers are 86dB/1W.. So I will need more than 10W. Let me take it 20W in average listening. 10,95v RMS into 6R makes 1,82A RMS and 2,56A peak.. Thats my current expectation.

What average voltage drop do you expect?
- For the regulators, 30 x 1,41= 42,3 minus diode drops makes ~41v. I will fix the outputs to 33v, so the voltage drop will be 8v on regulators. With 1,82A ~ 15W will be dissipated by the regulators in "average listening"

Have you calculated the heatsink and operating temperatures of the regulators?
- No. But I will use 0,5°C/W heatsinks for both one LM3875 and one regulator pair. For 20W average listening, LM3875 will dissipate ~13W and the regulators will ~15W then total thermal load of that heatsink will be ~45W (with looses). So, if I take 30°C (worst cases in home and in hot summer days) environment temperature then surface will be 30+(45x0,5)=52,5°C for average summerday listening.

What if the mains runs at near maximum tolerances for more than a few minutes?
- I am praying for this will not happen. But may be I can put a simple varistor+a fast fuse to mains input.

Can you explain why you think you need 375VA?
- For one channel I will need 3,25A RMS.. (sqrt(62x6))/6=3,23A.. 60 x 3,23 x 2 ~ 387VA.. Ok a bit more. But my first calculation was based on 30v rails. So this much error will acceptable. And peak values are not in calculation.. Why? I dont know...

If you do need 375VA then what ripple do you expect across the smoothing caps, C1 & 2?
- I dont know.. But 10,000uF can handle 4A at least, as I know.

What is the drop out voltage of the regulators?
- Calculated above..

What is the minimum mains supply voltage?
- Depends... But I havent seen any voltage under 200V until now.

Have you calculated if the 30Vac can give the required minimum voltage at the regulator inputs for worst case conditions?
- If ripples will not be calculated. The regulator (LT1083) will drop 1,25v.. So, to have 33v output I will need 34,25v at inputz. If mains drops down to 200v that makes 13% drift. Then the outputs will drop to 26,1... That makes 35,8v after rectifying, still enough..
 
You don't need a 960VA transformer. That's what the capacitors in the power supply are for. The capacitors serve a dual purpose, firstly as smoothing capacitors to reduce ripple, and secondly as energy storing devices to supply the high current transients the amplifier needs.

If your amplifier is going to be 2x 60W RMS then one of your 225VA transformers should be plenty.

Now, since you are using regulators, which will not be capable of supplying the high current transients, you'll need to ensure the capacitors on the output side of the regulators are large enough.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.