In the beginning .....

[gpsmithii]

Hello all I'm in the process of completing my first DYI project. The BOZ ....... will post pictures when I'm finished. I purchased the PCB for the original ZEN too and have just gotten around to making a commitment to get my tinkering lab in place again. I would like to build a 25 watt pr/ch version of the original. I have done some rough math (very rough) and determined that I would have to have a rail of +56 volts. A transformer of 600 VA with dual 40 volt secondaries should do the job. Correct? With exception to the needs lots-o-heatsink syndrome are there any other challenges I should consider? Thanks inadvance for your help.

[Nelson Pass]

Not quite. You are taking the AC secondary voltage ratings
and multiplying by 1.4 to get 56 volts, I assume.

First, you will lose a volt through the bridge. Then you will
lose another volt through ripple (typically)

Last, the primary / core/ and secondary losses will contribute
additional drop, depending on their ratings. Assume the
worst. Sometimes the losses can be quite a lot, depending
mostly on how conservative the manufacturer is and how
close you come to the ratings.

In your case I predict 50 volts if you use only one 40 V
secondary per channel. This is quite a bit higher than
the original Zen, but that doesn't mean you can't make it
work. Just get some really big sinks.

[gpsmithii]

Mr. Pass,

Thanks for your input. I did not give those losses any consideration. I'll order up a transformer with dual 44 volt secondaries.

[yldouright]

Nelson Pass

While we're on the topic of transformer losses, what would be a typical loss to expect from a frame transformer that has 8 taps from its screw terminals. Each wire has spade terminals and the wire resistance measured at the terminals is less than a milliohm.

[gpsmithii]

Would like to post some pictures of my Bride ... not sure how to do this using files local to my system. Help!

[GRollins]

yldouright,
The losses will depend on how heavily loaded the transformer is. Transformer secondary voltages can sag considerably if you draw a lot of current. Lightly loaded, the voltage will drop hardly at all. It depends on many factors--the resistance of the secondary, the inductance, how much core you've got and how permeable it is... Sure, if you had sufficient data about the transformer you could calculate the loss, but it's easier and faster just to hook it up and measure it.

Grey

[yldouright]

GRollins

Rightly said. Let me flesh out the parameters a little more so you can provide a more responsive reply. The transformer is a frame type with screw terminals and is rated at 3KVA that I am planning on using for an Aleph-X with an LC power supply. I am planning on loading the transformer with four amp channels, each drawing 8A at maximum power for a total of 32A which is significantly higher than the 24A I expect to get out of the transformer but since the rails will be in the 16-20V range and each channel will have 140mF of capacitance, I should be okay and clear of any sonic detriment. The transformer will have better than 5% regulation and so shouldn't sag very much under load but I don't know anything about permeability, which measure defines it or whether it is a sonic issue in an LC power supply. I guess what I was trying to determine was if the act of adding taps would in of itself create a problem or if it is just a matter of how many amps are being pulled from those taps. I expect the taps will be 25VAC under load and the bridges will take a volt off each tap as I understand it which will leave me around 22VDC after passing the coil less the internal losses of the transformer winding. I expect this to put me right in the 16-20V range on my voltage rails but in all my speculations, I am unsure so any feedback is appreciated.

[Panelhead]

Originally posted by yldouright
GRollins

Most transformers secondary voltage rating is at rated power. Since most are only run at a percentage of rated power the voltage I normally see is 150% of the rated voltage. This is using large high efficiency rectifiers and running the transformers at 40 - 50% of the rated power. The transformers are Plitron.
With little or no load I see 1.6 - 1.7 times. This is a good reason to use filter caps with a little higher than required voltage rating.
In a line level unit I built using a 18 volt trasformer the output was 31 volts. I had put 25 volt caps in and run it for a while before checking. I was really surprised to be getting that much. But the rectifiers used can make a volt or so difference also.

George



Rightly said. Let me flesh out the parameters a little more so you can provide a more responsive reply. The transformer is a frame type with screw terminals and is rated at 3KVA that I am planning on using for an Aleph-X with an LC power supply. I am planning on loading the transformer with four amp channels, each drawing 8A at maximum power for a total of 32A which is significantly higher than the 24A I expect to get out of the transformer but since the rails will be in the 16-20V range and each channel will have 140mF of capacitance, I should be okay and clear of any sonic detriment. The transformer will have better than 5% regulation and so shouldn't sag very much under load but I don't know anything about permeability, which measure defines it or whether it is a sonic issue in an LC power supply. I guess what I was trying to determine was if the act of adding taps would in of itself create a problem or if it is just a matter of how many amps are being pulled from those taps. I expect the taps will be 25VAC under load and the bridges will take a volt off each tap as I understand it which will leave me around 22VDC after passing the coil less the internal losses of the transformer winding. I expect this to put me right in the 16-20V range on my voltage rails but in all my speculations, I am unsure so any feedback is appreciated.

[GRollins]

Panelhead,
It's normal to see 1.414 x the incoming AC (ignoring diode losses for the moment) in a capacitor-input filter power supply. Nuthin' to do with the quality--or lack of it--of the power transformer. Essentially, the cap charges to the peak value of the AC instead of the RMS value that a voltmeter reads. That's the way a cap-input works. L-input filters behave differently. Diodes drop about the same voltage regardless of their other characteristics.
By my standards, if someone is drawing so hard on a power supply that it's collapsing all the way back down to a voltage equivalent to the incoming AC, it's waaaay too heavily loaded. But that's just me.
As for getting as much as 1.6 to 1.7 times the rated AC, Plitron builds in an extra volt or two on the secondary. I've got some 32VAC Plitrons that measure more like 34VAC. Some manufacturers do this, some don't. I happen to like Plitron transformers, myself, but it's a good idea to be aware that a Plitron transformer may give a different voltage in circuit than another manufacturer's transformer, even though they're rated the same.
yldouright,
We've just developed a minor crisis here at work...I'll try to get back in a bit...

Grey

[JOE DIRT®]

rated voltages of transformers are so eratic when not under any load especially when rectified from my experience. I have bench tested each transformer under a load test to record its behaviour before utilising it in a design and 5% variance is not uncommon between similar transformers.