If you have a nice transformer with a good VA rating, and your chosen filter capacitors but your +/- rail-voltages are a bit too high >
there is a simple low cost way to 'trim down' your voltage using series power diodes. You simply connect series diodes between your
bridge rectifier Plus & Minus outputs and your filter capacitors. Each diode reduces your voltage by .6V [ or very close to ] .
This means that 10x diodes in series with your + , and 10x diodes in series with your - , will reduce your rail voltages by 5.8 volts.
There is no limit on how far you want to go. The 'voltage regulation' is quite acceptible for most applications and as I said > it is low cost.
https://www.ebay.com.au/itm/4054292...-aedf-969fdf8decb6|iid:1|vlpname:vlp_homepage
I am likely to use this method in the design of an amplifier I am thinking of building. [ I got a great price for a high VA transformer, but its voltage is a bit too high ]
there is a simple low cost way to 'trim down' your voltage using series power diodes. You simply connect series diodes between your
bridge rectifier Plus & Minus outputs and your filter capacitors. Each diode reduces your voltage by .6V [ or very close to ] .
This means that 10x diodes in series with your + , and 10x diodes in series with your - , will reduce your rail voltages by 5.8 volts.
There is no limit on how far you want to go. The 'voltage regulation' is quite acceptible for most applications and as I said > it is low cost.
https://www.ebay.com.au/itm/4054292...-aedf-969fdf8decb6|iid:1|vlpname:vlp_homepage
I am likely to use this method in the design of an amplifier I am thinking of building. [ I got a great price for a high VA transformer, but its voltage is a bit too high ]
Not a great idea. Diodes aren't great packages to dissipate heat in. I've seen failures in commercial power supplies because they decided to use this idea.
If you've got volts to spare then it's better to fit +/- voltage regulators or power ripple eaters.
If you've got volts to spare then it's better to fit +/- voltage regulators or power ripple eaters.
As you should have noticed in the 'link' I provided, the method is to use HIGHLY OVERRATED DIODES 😎
[ Ohms Law indicates the heat in Watts @ .6V x Amps = no problem ]
[ Ohms Law indicates the heat in Watts @ .6V x Amps = no problem ]
Both expensive and more complex at high voltages > And this thread is "Simple & Low cost".
If you insist on this approach, then at least get some diodes you can bolt to a heatsink.
Forward voltage drop depends on the current through the diode, you should look at the diagram for that particular diode. If the amplifier is class AB, then the current will be variable, so the voltage drop on the filter capacitors at higher currents will be higher with those diodes than in the case of only diode bridge + caps. I don't know how it affects the sound of the amp. Certainly not positively.
As I said > 'Voltage regulation is quite respectible and decent' for common/regular use aplications.
OF COURSE the voltage drop across the diodes increases slightly upon higher current draw >
but as I stated, the diodes come before the filter capacitors and as such, with a good high-capacity 'bank of filter caps'
the 'effective regulation' is very good.
We don't listen to amplifiers at FULL CONTINIOUS POWER, so, your decent filter caps. serve very well for transient power output.
PS.
Rail 'voltage sag' is one of the most common situations associated with near ALL amplifier designs > upon full power output 😎
OF COURSE the voltage drop across the diodes increases slightly upon higher current draw >
but as I stated, the diodes come before the filter capacitors and as such, with a good high-capacity 'bank of filter caps'
the 'effective regulation' is very good.
We don't listen to amplifiers at FULL CONTINIOUS POWER, so, your decent filter caps. serve very well for transient power output.
PS.
Rail 'voltage sag' is one of the most common situations associated with near ALL amplifier designs > upon full power output 😎
As already mentioned by others above, diodes are a bad idea. If you must drop the voltage, consider using switching regulation schemes especially if the current is high.
A simple scheme is a magnetic amplifier:
https://www.ti.com/lit/ml/slup129/slup129.pdf
A simple scheme is a magnetic amplifier:
https://www.ti.com/lit/ml/slup129/slup129.pdf
Your post & link refers to a PWM power supply >
Once again, COSTLY & COMPLEX.
This thread and its title refers to "SIMPLE & CHEAP"
and applies to standard transformers + bridge rectifiers.
Once again, COSTLY & COMPLEX.
This thread and its title refers to "SIMPLE & CHEAP"
and applies to standard transformers + bridge rectifiers.
I would love to tell you in full detail ... but now is not exactly the right time.
What I can tell you is that it is an idea of an amplifier that has an audio transformer to drive the output stage.
The idea is completely 'driven by quriosity' being an amplifier that has by 'normal standards' > high harmonic distortion.
In my whole life, I have never listened to an amplifier that exhibits this > like 'old time' low power tube AND semi amps.
One of the primary goals is to have high power output > thereby removing the SUPER HIGH DISTORTION of clipping.
The drive to the audio transformer will determine its output clipping voltage > which is less Peak to Peak than the 'rail voltages'.
This is where I think I might reduce the rail voltages > either way, the output transistors will never 'fully saturate' to FULL ON.
The small reduction of 'rail voltage' could help the output transistors regarding SOA & heat. I plan for substantial heat-sink cooling.
The approximate power output of the amplifier will be 220 Watts per ch @ 8 ohms.
What I can tell you is that it is an idea of an amplifier that has an audio transformer to drive the output stage.
The idea is completely 'driven by quriosity' being an amplifier that has by 'normal standards' > high harmonic distortion.
In my whole life, I have never listened to an amplifier that exhibits this > like 'old time' low power tube AND semi amps.
One of the primary goals is to have high power output > thereby removing the SUPER HIGH DISTORTION of clipping.
The drive to the audio transformer will determine its output clipping voltage > which is less Peak to Peak than the 'rail voltages'.
This is where I think I might reduce the rail voltages > either way, the output transistors will never 'fully saturate' to FULL ON.
The small reduction of 'rail voltage' could help the output transistors regarding SOA & heat. I plan for substantial heat-sink cooling.
The approximate power output of the amplifier will be 220 Watts per ch @ 8 ohms.
I'm only asking to see the voltages, power and so on. I won't go into too much detail. I thought it was a lower power amplifier, based on your link for those plastic diodes. For such powers, you don't need those small 10A diodes, but some strong in TO220 case, so you can cool them. Since it's a serious design, I'd skip those diodes and get a proper transformer. Output transformers will also cost something (they will be big ones), capacitors, heatsinks ... there is no need to save money on the mains transformers for a project like this.
Last edited:
I've got a 600 VA transformer that I have connected via a 10 ohm primary 'current in-rush/surge' protection device.
With a simple bridge rectifier & filter capacitors, I have rail voltages of +/- 49 volts.
With a simple bridge rectifier & filter capacitors, I have rail voltages of +/- 49 volts.
Well, that's not much, not near enough for 220W/8ohm. That's it for 90-100W/8ohm with BJT output transistors. In my amplifier it is +-61VDC for about 140W/8ohm with BJT transistors. For mosfets you need +-5V more for the same power.
10ohm is small for inrush current limiting of such a strong transformer. At 230VAC I would put a 50ohm 50W resistor.
10ohm is small for inrush current limiting of such a strong transformer. At 230VAC I would put a 50ohm 50W resistor.
Last edited:
I am not using a simple resistor for 'in-rush' protection > I am using a proper in-rush protection device.
https://www.google.com/search?q=pea...wQyLTEzoAeUhwGyBwQyLTEzuAeIFw&sclient=gws-wiz
SO > If +/- 49 volts = 98 volts Peak to Peak
then RMS = basically 69 volts.
69 x 69 = 4761 > divided by 8 ohms = 595 WATTS.
SO > If +/- 49 volts = 98 volts Peak to Peak
then RMS = basically 69 volts.
69 x 69 = 4761 > divided by 8 ohms = 595 WATTS.
No, the RMS is 34.6V for 98V PP, but you will have a voltage drop on the transformer and filter capacitors and the amplifier cannot approach exactly the rail voltage, but a couple of volts less. So 100W RMS/8ohm is some practical max. for unstabilized power supply, probably a little less. Acoustically, the power doubling (100W-200W) is only +3dB, just enough to notice. If you want more power with that supply voltage, you have to go to bridge mode. That 600VA transformer is quite OK for 2x100W/8ohm and let's say 2x150W/4ohm.
"The peak-to-peak value is twice the peak value or 2.828 times the root-mean-square (RMS) value."
98/2.828=34.6
If you are building an amplifier with an output transformer, it will be really big if you want good bass. A lot of iron is needed for 20Hz or lower.
"The peak-to-peak value is twice the peak value or 2.828 times the root-mean-square (RMS) value."
98/2.828=34.6
If you are building an amplifier with an output transformer, it will be really big if you want good bass. A lot of iron is needed for 20Hz or lower.
Last edited: