I was about to remove it, but keep finding there is more to everything.
I very much appreciate the help I get as a noob.
I am fairly confident on this one but would love some clarification before I melt my psu.
I have a TPA3255 amp.
It came with an Aliexpress power supply.
I am pretty sure it is this:
https://www.aliexpress.com/item/100...8;-1;GBP+0.01@salePrice;GBP;search-mainSearch
States on the side:
S-800-48D
Output DC 48V 16.7A
I have plugged a watt meter into the mains socket and it is only using 16W.
Am I right in thinking that at 16W, 48V, it is only using 333.33mA?
In which case well under its limit.
It has a NOISY FAN. (not ideal when using for an amp eh?)
Am I right in thinking that I can remove the fan as it is well below its limits or is there something else I need to consider?
Thanks.
I very much appreciate the help I get as a noob.
I am fairly confident on this one but would love some clarification before I melt my psu.
I have a TPA3255 amp.
It came with an Aliexpress power supply.
I am pretty sure it is this:
https://www.aliexpress.com/item/100...8;-1;GBP+0.01@salePrice;GBP;search-mainSearch
States on the side:
S-800-48D
Output DC 48V 16.7A
I have plugged a watt meter into the mains socket and it is only using 16W.
Am I right in thinking that at 16W, 48V, it is only using 333.33mA?
In which case well under its limit.
It has a NOISY FAN. (not ideal when using for an amp eh?)
Am I right in thinking that I can remove the fan as it is well below its limits or is there something else I need to consider?
Thanks.
TEST. And consider, every 10 C rise on the die is half life. So, you could run it hotter, but for how long? As you have no capability to either calculate the internal die temp, or measure it, a heat sink should never be more than a bit warm to the touch. That is your only guide.
I have put in slower fans when I know I was way below the limits. It does not take much airflow to make quite a difference.
I have put in slower fans when I know I was way below the limits. It does not take much airflow to make quite a difference.
Get a quiet fan (db noise ratings are available) and put a thermal switch on it so it only turns on when needed. You can also go so far as to use a two-speed fan with two thermal sensors.
Or a DC fan with a linear controller. Might look at computer CPU fans and controllers.
In one big PA amp used for a sub, I took out the tiny high speed fan and fitted a larger low speed fan, added some cardboard ductwork to the airflow was where it was needed. If putting in a sensor, put a LED on it for "over temp" as fan bearings do not last forever and if it is quiet, the only telltale may be smoke.
16W is 16W. That is quite a bit.
In one big PA amp used for a sub, I took out the tiny high speed fan and fitted a larger low speed fan, added some cardboard ductwork to the airflow was where it was needed. If putting in a sensor, put a LED on it for "over temp" as fan bearings do not last forever and if it is quiet, the only telltale may be smoke.
16W is 16W. That is quite a bit.
OK, thank you gents. I am glad I asked.
I kind of thought because it is only running at 2%, it would be fine, my thinking was comparing to a 60W psu sitting here that doesn't have a fan at all.
Sounds like it is not worth the risk though.
I shall go rummage, I think I have a couple of larger pc fans.
I am aware that larger fans run slower move as much air and are quieter.
I like the idea of a temperature controller or thermal switch for in the future.
I shall look into how to do that, thanks.
On a side note, would you say to replace this power supply?
Seems like a bit overkill to me.
Or is it appropriate?
I kind of thought because it is only running at 2%, it would be fine, my thinking was comparing to a 60W psu sitting here that doesn't have a fan at all.
Sounds like it is not worth the risk though.
I shall go rummage, I think I have a couple of larger pc fans.
I am aware that larger fans run slower move as much air and are quieter.
I like the idea of a temperature controller or thermal switch for in the future.
I shall look into how to do that, thanks.
On a side note, would you say to replace this power supply?
Seems like a bit overkill to me.
Or is it appropriate?
You were looking at an idle current. Your amp is rated as 300 + 300, so into the lowest safe impedance, what is the peak draw? A lot more! No watt-meter is going to show you that, only a current probe on a storage scope. The question is, how clean is it? Those are made mostly for digital systems, not audio. A little searching and you can find post-switcher filters to clean them up. I am still a fan of linear supplies, but I am old fart and just have not learned the issues with switchers for audio, only digital. It is not the technology, but the implementation. Usually. A linear supply for a 300W amp is BIG and EXPENCIVE.
Ah, thanks, I see what you are saying.
I was playing music at the time, but I do see what you are saying.
I was hoping to just actually use it while I start to ponder that same question.
Currently it is SO loud it is unusable.
Thank you for the words 'post-switcher filter' by the way.
Decision to make is whether to spend time sorting out an appropriate fan or sell this one and get a 'better' power supply.
I am very much a noob and have got used to 'linear equals good' and then suddenly i was reading a bit about maybe not so good .... lol, one can't generalise it seems.
"A linear supply for a 300W amp is BIG and EXPENSIVE." Ah, that seems like you have helped me make my decision for now 🙂
It is one of these:
https://www.jlelectronicsph.com/pro...filter-feedback-ready-to-run-amplifier-moduleSo I shall pile into the appropriate thread to see where they got to with power supplies.
With this new great search function it will be a lot simpler.
I was playing music at the time, but I do see what you are saying.
I was hoping to just actually use it while I start to ponder that same question.
Currently it is SO loud it is unusable.
Thank you for the words 'post-switcher filter' by the way.
Decision to make is whether to spend time sorting out an appropriate fan or sell this one and get a 'better' power supply.
I am very much a noob and have got used to 'linear equals good' and then suddenly i was reading a bit about maybe not so good .... lol, one can't generalise it seems.
"A linear supply for a 300W amp is BIG and EXPENSIVE." Ah, that seems like you have helped me make my decision for now 🙂
It is one of these:
https://www.jlelectronicsph.com/pro...filter-feedback-ready-to-run-amplifier-moduleSo I shall pile into the appropriate thread to see where they got to with power supplies.
With this new great search function it will be a lot simpler.
So, an 800 Watt supply, which came-with the 600W audio power amp.
At idle or very soft playing, yes, 2% of maximum power is reasonable. 1 Watt of audio is fairly loud. It might not get warm with the fan, only slightly warm without the fan.
Now go away. Your dogs will have a party and play the stereo REAL LOUD so they can sing-along. (I had friends and customers like that.) The hardly-warm (@1W) amp will get hot as heck at 100W-'500W', and might burn-up.
It is difficult to realistically test an audio amp (and power supply) for "hard abuse" like this. (Using it in a rock band is one way.)
If you don't have dogs or drunken users, you can play it while testing your hand on it every few minutes. That works OK to a point. In traditional amplifiers that "last few dB" was sometimes the difference between "OK" and YOWW!!". (And sometimes a few dB more, heavy distortion, was actually cooler.)
What I'd actually do: peek inside and get the fan specs. Say 48V 0.05A. That's equivalent to a 960Ω resistance. Get a 1kΩ resistor, rated 2 Watts, and splice it into the fan lead. That will cut the fan down to roughly half voltage and (if a DC fan!) very roughly half-speed, which may be quarter noise. Which is often enough to be tolerable.
No. It consumes 70 mA at 230V mains voltage which is 16W including losses. But... was this with a load or without? And what will the load be?
If your load consumes 50W for 2 channels omitting the fan will not work out OK. Engineers, even those in the Far East, don't use a fan without a reason. Now, things can be done passive but then one would need a lot of surface and much aluminium with large cooling fins. Metal = cool = good = expensive = heavy = higher shipping = unwanted. A fan is cheaper which appeals to a lot of people. In general the least qualitative and also cheapest solutions are liked best.
What can be done is to have the fan only operating when the PSU delivers a certain power to the load. Or... try out a design based on MA12070 and use a simple linear PSU with enough heatsink. Feed your speakers 5W maximum (likely around 1W in real life) and enjoy fanless audio.
Last edited:
You can use a lower noise fan.
Which as you mentioned can be a larger fan at lower speed.
Or speed control the existing fan.
You can probably find PWM speed controller which enough current for around 6 to 12 dollars.
Even though the amplifier is capable of doing around 220 watts per channel 4 ohms
likely either 1% or 10% rating...not sure
Point is your actual needed amplifier output power might be as low as 10 watts maybe 30 watts.
So the recommended minimum supply is only 200 watts.
there is fan less power supply. then also can add low speed fan
to improve lifespan.
3255 can operate from 18 to 52 volts
so if your power demand is not high
you can likely use the 60 watt supply.
Which as you mentioned can be a larger fan at lower speed.
Or speed control the existing fan.
You can probably find PWM speed controller which enough current for around 6 to 12 dollars.
Even though the amplifier is capable of doing around 220 watts per channel 4 ohms
likely either 1% or 10% rating...not sure
Point is your actual needed amplifier output power might be as low as 10 watts maybe 30 watts.
So the recommended minimum supply is only 200 watts.
there is fan less power supply. then also can add low speed fan
to improve lifespan.
3255 can operate from 18 to 52 volts
so if your power demand is not high
you can likely use the 60 watt supply.
Now that sounds right but ... 60W can be let's say 15V at 4A (good!) or 30V at 2A.. it can also be 60V at 1A. Or 120V at 0.5A 🙂
Now maybe it is wise to have low(er) voltage designs that can deliver Amperes to loads like loudspeakers. The obvious trend is very efficient lightweight (no heatsinks) extremely high power class D or similar devices fed by equally lightweight high power SMPS devices that can deliver that power and therefor need to be cooled actively. They can deliver that power but will they in a home setup? What would you do if you were a designer? Design for what it can deliver or for what it may deliver in specific cases?
The other way of thinking could be to use devices tailor made for the situation with a certain broad margin. Or to adapt the design to a lower voltage like 20V 5A which can be done linear without much trouble. "New" possibilities will arise and things will be more easy manageable. Also more silent.
Now maybe it is wise to have low(er) voltage designs that can deliver Amperes to loads like loudspeakers. The obvious trend is very efficient lightweight (no heatsinks) extremely high power class D or similar devices fed by equally lightweight high power SMPS devices that can deliver that power and therefor need to be cooled actively. They can deliver that power but will they in a home setup? What would you do if you were a designer? Design for what it can deliver or for what it may deliver in specific cases?
The other way of thinking could be to use devices tailor made for the situation with a certain broad margin. Or to adapt the design to a lower voltage like 20V 5A which can be done linear without much trouble. "New" possibilities will arise and things will be more easy manageable. Also more silent.
Last edited:
Yeah that is the ironic thing about switching amplifiers and power supplies.
Efficiency is very high, so it makes it very easy to use fan less design.
But to meet price point the heatsink is reduced and still need fan.
When incredible basic linear supply, and larger amplifier heatsink.
would make quiet amplifier.
Or switching supply with larger heatsink can be fan less
looking at size of amplifier heatsink, with 200 watt power supply, likely easy
to shut it down thermally.
Not talking down the amplifier, when they say. Minimum supply 200 watts.
It should read, Maximum supply 200watts
400 watts with forced air cooling.
Most likely non issue since most are only using 18 to 30 watt power.
per channel. for average to medium listening levels, or actual speaker distortion
Efficiency is very high, so it makes it very easy to use fan less design.
But to meet price point the heatsink is reduced and still need fan.
When incredible basic linear supply, and larger amplifier heatsink.
would make quiet amplifier.
Or switching supply with larger heatsink can be fan less
looking at size of amplifier heatsink, with 200 watt power supply, likely easy
to shut it down thermally.
Not talking down the amplifier, when they say. Minimum supply 200 watts.
It should read, Maximum supply 200watts
400 watts with forced air cooling.
Most likely non issue since most are only using 18 to 30 watt power.
per channel. for average to medium listening levels, or actual speaker distortion
There is so much wrong with your assumptions that I don´t even know where to start.
I´ll try:
1) you won´t melt it, it will catch fire.
2) as mentioned above, Watt Meters are crude devices, best case catch average consumption.
3) WRONG, you are mixing 48VDC current capability (up to 16.7A) with MAINS 230VAC consumption, what your Watt meter is (poorly) measuring. I feel I am wasting my time here.
4) you have a 48*16.7 800W output power supply.
Even if 90% efficient, it will have to dissipate whopping 80W heat.
For reference, put an incandescent 80W bulb inside a closed cardboard box and turn it on.
Save some time by dialing 911 for the firefighters NOW, they will take at least 10-5 minutes to arrive and 10 more setting up the water lines, etc.
Hint: there is a reason manufacturers put that fan there.
5) those suggesting OP mods his PSU, add a variable speed controller, even a thermal switch, are very optimistic about his qualifications.
His very questions do not give me much hope.
And I´m being polite, as per Forum rules.
6) those suggesting he permanently blocks fan speed, without means to thermally bypass that when needed, should contribute to his fire insurance prime.
7) to the OP: leave as is, or get an Audio rated (hint:less a noise, let alone less electrical noise) power supply.
Cheap is not best here.
EDIT:
Again: in principle, I agree, just way above the OP.
Not to far above.
OP already figured out larger low speed fan
or even questioned the need for such a large supply.
and going with a fan less design.
All good assumption.
measurement was taken on line side.
higher voltage/ lower current.
its just misleading or surprising since at listening
levels power usage was low.
its why people ask questions.
think it was missed
The supply is providing lower voltage.
so that is where the higher current is.
OP already figured out larger low speed fan
or even questioned the need for such a large supply.
and going with a fan less design.
All good assumption.
measurement was taken on line side.
higher voltage/ lower current.
its just misleading or surprising since at listening
levels power usage was low.
its why people ask questions.
think it was missed
The supply is providing lower voltage.
so that is where the higher current is.
Apologies for disappearing there gentlemen.
Life. You know.
Thank you all for your input.
OK, so, the simple answer is no. Do not remove the fan. I am glad I asked.
@JMFahey
You are correct, I have very little knowledge and so am being very careful at each stage.
Hence me asking a question before I do anything that I can’t find the answer to.
I thank you for your detailed response. (and your restraint).
The only thing I would disagree with is that it was a waste of time.
I learned a lot from people discussing my options.
I am keen to learn how to do these things.
First task is always finding out what can be done and the terminology used to be able to look into how to do a thing.
This psu arrived with the amp but only came with it from someone else, it wasn’t produced with it.
https://www.jlelectronicsph.com/pro...filter-feedback-ready-to-run-amplifier-module
So, the aim was to just use it for now while I start to understand how to go about getting a worthy psu for it.
I have listened to the advice and kept a fan on it.
I have tried running it and feeling the heatsink and it felt reasonable at the levels I was using it at.
I am also very aware of my limitations and I know that I definitely don’t know what I don’t know yet.
So, with that in mind, I dug out an old Noctua pc fan. (both it and the original are 12V)
To give an idea, on the bench, measured with a phone app (not aiming for accuracy, just getting a rough idea) 2-3 inches from the phone with the air flow away from the phone.
The old one was 69dB and the Noctua 54dB.
I have attached that to the top of the psu and drilled a load of holes in the cover.
It is still noisy enough to be distracting and I am finding that I am not using it.
I am very happy to test and measure, but I don’t know how to do so.
I do plan on getting a more suitable psu when I understand what ‘better’ is when it comes to a SMPS.
I believe that the Meanwell uhp 200r-48 is mentioned as being acceptable on the relevant thread.
This is fanless, but I have yet to learn about psu electronic noise.
I get the general gist of LPSU and ripple but not SMPS yet.
Also, the amp itself has a linear regulator so I don’t know to what degree the amp is doing this and so how far do I need to go with the power supply.
I am looking forward to being able to measure such things as ripple, but do not have the equipment yet.
In the mean time, I am looking for a way to make this psu (unless I just buy another now based on advice) more usable as it is what I have.
It seems to me like my Noctua fan is producing better airflow than the smaller one, but I also can’t test that.
However, it would seem to me that I could get away with slowing it down somewhat.
I don’t know yet how to add a variable speed controller or a thermal switch.
Having read those words, I could look them up.
I am open to advice about whether it is beyond me.
I also like the idea of fail safes everywhere.
I am, as, JMFahey stated, very much a noob and am concerned about doing things that are wrong as a lot of things that get suggested assume an amount of knowledge and so I might miss something.
As said, perhaps a new psu is the better option given my current levels.
Life. You know.
Thank you all for your input.
OK, so, the simple answer is no. Do not remove the fan. I am glad I asked.
@JMFahey
You are correct, I have very little knowledge and so am being very careful at each stage.
Hence me asking a question before I do anything that I can’t find the answer to.
I thank you for your detailed response. (and your restraint).
The only thing I would disagree with is that it was a waste of time.
I learned a lot from people discussing my options.
I am keen to learn how to do these things.
First task is always finding out what can be done and the terminology used to be able to look into how to do a thing.
This psu arrived with the amp but only came with it from someone else, it wasn’t produced with it.
https://www.jlelectronicsph.com/pro...filter-feedback-ready-to-run-amplifier-module
So, the aim was to just use it for now while I start to understand how to go about getting a worthy psu for it.
I have listened to the advice and kept a fan on it.
I have tried running it and feeling the heatsink and it felt reasonable at the levels I was using it at.
I am also very aware of my limitations and I know that I definitely don’t know what I don’t know yet.
So, with that in mind, I dug out an old Noctua pc fan. (both it and the original are 12V)
To give an idea, on the bench, measured with a phone app (not aiming for accuracy, just getting a rough idea) 2-3 inches from the phone with the air flow away from the phone.
The old one was 69dB and the Noctua 54dB.
I have attached that to the top of the psu and drilled a load of holes in the cover.
It is still noisy enough to be distracting and I am finding that I am not using it.
I am very happy to test and measure, but I don’t know how to do so.
I do plan on getting a more suitable psu when I understand what ‘better’ is when it comes to a SMPS.
I believe that the Meanwell uhp 200r-48 is mentioned as being acceptable on the relevant thread.
This is fanless, but I have yet to learn about psu electronic noise.
I get the general gist of LPSU and ripple but not SMPS yet.
Also, the amp itself has a linear regulator so I don’t know to what degree the amp is doing this and so how far do I need to go with the power supply.
I am looking forward to being able to measure such things as ripple, but do not have the equipment yet.
In the mean time, I am looking for a way to make this psu (unless I just buy another now based on advice) more usable as it is what I have.
It seems to me like my Noctua fan is producing better airflow than the smaller one, but I also can’t test that.
However, it would seem to me that I could get away with slowing it down somewhat.
I don’t know yet how to add a variable speed controller or a thermal switch.
Having read those words, I could look them up.
I am open to advice about whether it is beyond me.
I also like the idea of fail safes everywhere.
I am, as, JMFahey stated, very much a noob and am concerned about doing things that are wrong as a lot of things that get suggested assume an amount of knowledge and so I might miss something.
As said, perhaps a new psu is the better option given my current levels.
You could always put a normally open thermal switch in series with the fan to only turn on when hot.
I did not look too hard, this would work but you can probably find something cheaper.
https://www.digikey.com/en/products/detail/kemet/OHD3-70M/5409419
I did not look too hard, this would work but you can probably find something cheaper.
https://www.digikey.com/en/products/detail/kemet/OHD3-70M/5409419
Last edited:
OK, thanks.
You will be surprised to hear I wasn't aware of those 🙂
I assume that these are for the air temperature or attached to the heatsink.
Positioning would be important wouldn't it?
You will be surprised to hear I wasn't aware of those 🙂
I assume that these are for the air temperature or attached to the heatsink.
Positioning would be important wouldn't it?