Hello
Due to size constraints, I am looking at using 2x 250VA transformers for a particular two channel design and am wondering which option is theoretically more beneficial. The design is LMOSFET and most likely will be used driving a 4ohm impedance. I am looking at both 80VCT and 90VCT and realize that at 8ohm impedance, that the 90VCT option will produce more power, however I am uncertain about which option in practice will produce the best bang for the 4ohm impedance.
Given normal losses and transformer limitations, which would be the better choice?
Thx
Due to size constraints, I am looking at using 2x 250VA transformers for a particular two channel design and am wondering which option is theoretically more beneficial. The design is LMOSFET and most likely will be used driving a 4ohm impedance. I am looking at both 80VCT and 90VCT and realize that at 8ohm impedance, that the 90VCT option will produce more power, however I am uncertain about which option in practice will produce the best bang for the 4ohm impedance.
Given normal losses and transformer limitations, which would be the better choice?
Thx
I'd use this, but I'm unconventional.
500W +/-45V amplifier dual-voltage PSU audio amp switching power supply board | eBay
500W +/-45V amplifier dual-voltage PSU audio amp switching power supply board | eBay
Your numbers don't make any sense.
A 250VA transformwer is suitable for an amplifier producing a maximum output of 125W to 250W.
You have stated that you want 4ohms capability.
Take your 250W into 4ohms as a starter.
The maximum output required to hit that target is Vpk=sqrt(240*4*2) = 43.8Vpk
An 80Vac centre tapped transformer will give ~ +-58Vdc
That is far above the required Vpeak.
I got over 310W into 4r0 from a dual 40Vac transformer. For two channels I adopted a 1000VA transformer, not 2 times 250VA.
For 4 ohms Id almost always go with the lower voltage option , but not knowing anything about the amplifier and it's function* , is sorta like asking a blind man which color looks better. IMO you've already compromised the supply by using 2 transformers. A 500 VA transformer has much better regulation than 2x 250. Space wise the volumetric efficiency is better using one transformer too.
*( it could also mean the difference of not blowing up the output section, better SOA and all that wider swing of rail voltages loaded vs unloaded)
Last edited:
And moreso with lower load impedance.. And moreso with higher rail voltages. So
First I need to clarify..
My design is a 2 channel, so I am using 1x250va for each channel.
Lets assume 250w @4ohm
The current @ 250w per rail is around 2.8 amps
The 90vct trans is rated at 2.8A per rail
With losses considered, is it not more prudent to use the 90vct and gain power for 8ohm drivers as well? Losses considered? The basic math doesn't account for the losses.
Just seeking clarity
Maybe I am looking too closely
Switching power supply for audio amplifier
It is unconventional indeed. How much the +/-45V output drop when outputting 5.5 amp? Did you check the ripple of the output rail at idle and at full current out?
250W at 4ohms is 7.9A RMS. Half of this comes from each rail, and average is about 0.9 of RMS so the DC current draw will be 3.6A per rail.synonymous said:
A 250VA 90V winding can supply 2.8A RMS AC continuously. This may be about 1.5-2A continuous DC from a capacitor input PSU. It could supply up to about 3A on peaks, provided that the duty cycle was not too large (i.e. you are listening to real music with quiet and loud parts).
For 250W at 4 ohms you actually need something like 70V CT at 250VA. If your amplifier is inefficient then 80V at 300VA. On that basis given the choice between 80V and 90V you should choose 80V and accept that you won't get 250W into 4 ohms, except possibly on rare peaks.
Send them a message asking them. I have not used this one (I build tube amps) but I use similar SMPS boost converters ( 12V - 280V or 380V 150W or 500W )
Hmm certain clues tell me that amplifier is NOT designed for 4-3 ohm speaker duty.
Sure you can run it that way but you will have lower performance plus run the higher risk crowbarring the outputs and smoking speakers. Id recommend lowering the SOA by reducing the DC supply rails. Which off hand, is limited by the driver swing to turn the enhancement MOSFETs gates fully on. Ideally an amp like this has 2 extra HV rails. I reckon there is 15 V min. headroom on that design even at the higher ohm loads. That might be a good voltage amp, not so great for supplying large current peaks for low ohm loads.
Has anyone ran full power test data on 4 ohm loads. IMO The protection circuity looks crude. It might prevent fires but not speakers. You could actually use a sagging supply voltage to offer some form of protection on ill advised low ohm loading. Consider going lower than dual 250VA transformers.
Sure you can run it that way but you will have lower performance plus run the higher risk crowbarring the outputs and smoking speakers. Id recommend lowering the SOA by reducing the DC supply rails. Which off hand, is limited by the driver swing to turn the enhancement MOSFETs gates fully on. Ideally an amp like this has 2 extra HV rails. I reckon there is 15 V min. headroom on that design even at the higher ohm loads. That might be a good voltage amp, not so great for supplying large current peaks for low ohm loads.
Has anyone ran full power test data on 4 ohm loads. IMO The protection circuity looks crude. It might prevent fires but not speakers. You could actually use a sagging supply voltage to offer some form of protection on ill advised low ohm loading. Consider going lower than dual 250VA transformers.
Last edited:
No tests yet
These laterals don't require any fancy protection and are in spec
The crowbar is on the rails
The extra rails you speak of aren't going to be used in this design, however, what you mention is exactly what I am referring to, amongst other losses involved.
Given the losses involved, to maximize the performance using a 250va transformer with a 4ohm load, I have estimated a 84vct trans, however there's no 84vct option to choose from. This is why I'm inquiring.
The voltage to produce maximum continuous power at the output is greater than the basic calculation. It requires more voltage to the point of diminishing current. If we just look at it as like you are saying, an additional 15v per rail, that's
sqrt(250*2*4)=45 +15v=60v *.7=42+42=82vct.
Being in the middle, losses considered, more V or more A
Thanks
These laterals don't require any fancy protection and are in spec
The crowbar is on the rails
The extra rails you speak of aren't going to be used in this design, however, what you mention is exactly what I am referring to, amongst other losses involved.
Given the losses involved, to maximize the performance using a 250va transformer with a 4ohm load, I have estimated a 84vct trans, however there's no 84vct option to choose from. This is why I'm inquiring.
The voltage to produce maximum continuous power at the output is greater than the basic calculation. It requires more voltage to the point of diminishing current. If we just look at it as like you are saying, an additional 15v per rail, that's
sqrt(250*2*4)=45 +15v=60v *.7=42+42=82vct.
Being in the middle, losses considered, more V or more A
Thanks
that can't be right.
Ioutput into a 4r0 test load would be Iout=Sqrt(Pmax/Rload)
The peak output current is sqrt(2)*Iout = Sqrt(Pmax*2/Rload) = sqrt(250*2/4)
= 11.18Apk
Then account for the current into a reactive speaker load which can approach and even exceed three times the dummy resistive load. i.e. >33Apk
These output currents come from the supply rails.
Not 2.8A per rail
Thanks andy
I'm just looking for the most efficient use of a 250va transformer, continuous drive, not when lightning strikes or for any variety of other loads are attached, just a nominal 4ohm load.
Here's a snippet from a project on Rod Elliot page using the same output devices pairs..
"With the full supply voltage of +/-70V (which must not be exceeded!), continuous ('RMS') power is around 180W into 8 ohms, or 250W into 4ohms. Short term (or 'music') power is typically about 240W into 8 ohms and 380W into 4 ohms. Note that depends to a very great degree on the power supply, and a robust supply is a requirement for the maximum output. The recommended supply voltage is ±56V.
As noted, unless you really need the maximum possible power, I suggest that you use a supply voltage of ±56V obtained from a 40+40V transformer. You will get around 150W into 8 ohms from this supply voltage, but you also relax the demands placed on the MOSFETs and heatsinks. The difference between using ±56V and ±65V is less than 1dB - for the extra peace of mind and relaxed heatsink requirements that's a very small price to pay."
Based on Randy Slone...
70wrms @ 8ohm
(24x24)/8=72w
24x1.414=34v
At max power, it is estimated the power supply will droop 3-6v, so must be added
34+5=39
Since using mosfets, we must increase the voltage by 7% to compensate for losses
39x1.07=42v
Add for rectifier diode
42+1=43v
Back calculate for transformer
43x.707=32x32
Calculate current
24/8=3A or 1.5A each secondary
Assuming normal losses and OPS efficiency
70wrms @8ohm 110wrms @4ohm
According to the efficiencies, I can create in the range of 150wrms @4ohm using a 35+35, limited by voltage, and 170wrms @4ohm using a 40x40 250va, limited by current @3.1A /rail, and achieve 250wrms @4ohm using a 45x45 @4.6A /rail 400+va trans.
So in actuality it only makes sense to use the 40x40 in this case. The only other factor is the possibility of 8ohm, or if I can get them custom made.
70wrms @ 8ohm
(24x24)/8=72w
24x1.414=34v
At max power, it is estimated the power supply will droop 3-6v, so must be added
34+5=39
Since using mosfets, we must increase the voltage by 7% to compensate for losses
39x1.07=42v
Add for rectifier diode
42+1=43v
Back calculate for transformer
43x.707=32x32
Calculate current
24/8=3A or 1.5A each secondary
Assuming normal losses and OPS efficiency
70wrms @8ohm 110wrms @4ohm
According to the efficiencies, I can create in the range of 150wrms @4ohm using a 35+35, limited by voltage, and 170wrms @4ohm using a 40x40 250va, limited by current @3.1A /rail, and achieve 250wrms @4ohm using a 45x45 @4.6A /rail 400+va trans.
So in actuality it only makes sense to use the 40x40 in this case. The only other factor is the possibility of 8ohm, or if I can get them custom made.
Good luck finding an bay seller who understand technical question. I was hoping you can provide hands on experience to see if those PSU was really practical. And I have serious doubt.
250VA is good for about 220 Watts DC.
A great amp is 78% efficient at clipping. Real amps run under 70%; yours may be 65%.
65% of 220 Watts is 142 Watts. This is the maximum long-term safe power you can extract as Sine output.
142W in 4r is 67V p-p. Allowing 10% overage for losses, 73V total supply, +/-37VDC. This is about 2*24V AC.
That is for *continuous* power. Shaker-table work. Bench testing in hot resistors.
Speech/music has low duty cycle. I am sure you can have much higher "peak" power, accepting that it will sag like my power-line when I run the dryer. But going 1.4X higher voltage (double power) may get saggy on sustained crescendos. 2*33V may be as high as is safe for 4r and speech/music.
A great amp is 78% efficient at clipping. Real amps run under 70%; yours may be 65%.
65% of 220 Watts is 142 Watts. This is the maximum long-term safe power you can extract as Sine output.
142W in 4r is 67V p-p. Allowing 10% overage for losses, 73V total supply, +/-37VDC. This is about 2*24V AC.
That is for *continuous* power. Shaker-table work. Bench testing in hot resistors.
Speech/music has low duty cycle. I am sure you can have much higher "peak" power, accepting that it will sag like my power-line when I run the dryer. But going 1.4X higher voltage (double power) may get saggy on sustained crescendos. 2*33V may be as high as is safe for 4r and speech/music.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.