I have been watching your videos for few years and im very happy to see that you gotten into audio stuff, because i recently acquired same interest. Got myself amplifier and thinking about building speaker boxes for varies speakers i have laying around.
These Class T amps are impressively efficient. I've used one in my shop for years and it hardly heats up, which was great when I ran it on a solar-charged battery. Nice to see some real numbers from you on this, thanks!
I have the exact same amplifier and I was running into problems with clipping and also just an overall unpleasant sound using an inadequate power supply. I had a cheap 24v 20A switch mode power supply lying around (the type used for security cameras) so I hooked that up and was able to drive the amp much higher without the clipping and a significant improvement in the sound quality. The power supply only cost around $20 and so far it seems to work really well.
Okay, so I checked again. I actually have to plug it in to the amp first and then the wall after... Otherwise I will burn it in an instant. For now it works just fine with my pair of Tapco 6915 speakers. Does not handle the bass as well as my NAD, but it sounds really clean. Guess I just been Lucky 🙏👍 Thanks!
Could you please add a link to your benchtop power supply build video that you mention? I can't find it, but I need a really good benchtop power supply, and I can't afford to buy a new one that puts out what I want, so I'd be curious to see how you made yours!
I've got a few of these chinesium 3116 based amps at home. I find that a UL listed laptop power supply generally hits the sweet spot. 19V might be pushing it for some of these, you have to check the caps. I have a hifi and the curse of the golden ears but for 18 bucks plus a trip to the thrift its good enough for my workshop.
If you look more closely at the specs you'll see those power ratings are at 10% distortion. They make about 2/3 that max power with 1% distortion. There's another thing to consider with measuring this way, the frequency range of the meter. Most meters of that type have a rather low frequency limit for AC volts, meant for measuring household current. While difficult to get precision, using the scope is perhaps better. That's not to say a given meter won't measure 1kHz but you need to check the specs. Measure the input current while doing such a test. 10% distortion might not be all bad considering when listening to music that output is normally momentary. It's a peaks, not over long notes, generally. Distortion is the main thing with class D. Since the sound is *recreated* at output filtering a not processed at all in the linear region of the amplifying device there is virtually no noise. And specs say so. The signal to noise ratio is outstanding for such an amplifier.
Yeah, bought one of those chinese S.M.S.L-amps, looking very similar to the one you got there, rated at 2x50 Watts into 4 ohms, with a 24 VDC supply. What they ship it with however is a 12 VDC brick. Not a word about that in the description. Contacted them and they had some poor excuses but promised to send a better power supply. Some time later it actually showed up. 18 V. Hahaha. They didn't lie. It is a little better...
I have a similar amp based on the same chip. It measured at 32db gain, which is way too high.I lowered it to 20db, which resulted in a more linear volume control function and far less noise. I was actually quite surprised that it could drive my B&W Matrix 805s to decent levels.
You'll find that rating is for at 24v and they got it from the chip spec sheet. But they include an 18v power supply with too little of current supply. Sometimes you might find a 24v power supply included, but even that is often no good too since the rated current output is too low to provide the wattage the board is asking for at that voltage. You can get the right power supply for about £20 here in the UK, probably cheaper over in America. The way you calculate if you have enough is by getting the supply voltage and current that the supply is rated for, multiplying them and dividing them by the amps efficiency (e.g. /0.8). That tells you if it's capable of providing enough for your amp. In this case 24v is the max for the board, so get a higher current output capable one to get to that figure you need!
I have work with these class D amplifier from TI, they have one chip that is rated at 100W/ch with almost no heatsink! In continous you can not expected more than 10W/ch without a thermal cutout! Plus these chineese amp PS are underrated for the expected power. Maximum power is rated with distortion at 10%! You can't add any big heatsink...
Those TI based amplifiers are pretty nice (and cheap!). There are definitely some amps that cut corners, but a lot of them look pretty good. 50c says that a 19.5v is a common laptop power supply to keep things cheap, and that sourcing an acceptable 24-26v PSU would bump the cost a bit much.
Doin' My Part Brother👍 I think I would have liked to see the weak link power supply theory put to the test. (not at all mocking you, I just would like to see the difference)
You should've shown how/if it's power output improved when powering it with your big homebuilt benchtop power supply you showed there instead of that tiny brick power supply, as I'm sure your benchtop supply could actually deliver what that amp really wants in order to perform at its best.
Try to also measure the output of the power supply under load. Was the amp taxing the power supply? Plus, It would be interesting to see the efficiency of the amplifier.
When measuring AC power, the power is measured differently as we have a periodic voltage and current. So the true rms power is much less than what you measured as the equations apply for continuous dc voltage and current.
He measured AC volts in Vrms. He used the correct formula for power (Vrms^2 / R). I think you are somehow a bit mixed up about something. I am an electrical engineer, so I have studied this stuff and use it all the time. I am not mixed up.
I have a question. Back 25 years ago when I was ripping out my back seat and filling that area and hatchback with speakers, I remember them offering 4 ohm or 8 ohm on the speakers I purchased. My question to you is if it theoretically splits the power in half, why would you want to go with 8 ohm speakers over 4? Or is it you want to match the speakers to the amp? Like I said before, I crammed 4 12's and 3 15's in the back of an eclipse hatchback, so sound quality wasn't as important as shaking grandma's fine China off of her display shelves...
I learned alot over the years in the field I was playing in. I was in what was known as decibel drags. Basically pull up your car, they stick a decibel meter in and you see how high you can register. I learned how far the sound waves need to linda bounce around in the car before hitting their peak, the angles the speakers needed to be to obtain that and temperature of the air in the car for best results. But as I stated before, quality of the sound wasn't extremely important, just more loud the better.
An 8 ohm load is easier for an amp to drive, especially a low cost one. An amp will try to supply double the current into a 4 ohm speaker but it needs a larger power supply (transformer, usually) to do that.
This is why a single slot car stereo only gets max 40W per channel. To get more, an internal power supply to step up the voltage is required, but all this does not fit into a normal slot, let alone the necessary heat sinks.
Can you test the power supply to see how much it's max output is? That would be useful to know. Also, you need to look at the spec sheet for the amp and see what the specs really say, on good quality amps it will say "50w @ 4 Ω at .01% THD" but on cheap amps, it will say "50w @ 4 Ω at 10% THD" which means essentially a square wave and 1000 times more distortion than the first amp. You have to really read the specs and hope they aren't lying. Some cheap amps will just outright lie on the spec sheet.
I used to laugh at those flea market amps that had 300 Watts on the box. You just know they are measuring peak to peak instead of RMS (.7 x p-p/2) values. Then try pointing it out the the vendor who wouldn't know a volt from an ohm.
Electronics. Course on makers mob? It's common, when u buy an amp ... That the weak point is the power supply, so the amp doesn't have as much output as it could?
Beware running any amp continuously at its nominal power rating - even 'good' ones will overstate what they can really survive continuously and I have popped an amp doing exactly this kind of test.
Actually not the case with this board. I've tested it full load 24v (with a heatsink of course, that's needed at 24v) for an entire 24hrs with no problems
You're not going to hear much difference in loudness between 35 and 50W. To sound "twice as loud" you need 10x the power. Most people can start to just perceive a difference in loudness at 1.5 dB(SPL). That's 1.4x the increase in power. And that's precisely the increase between 35 and 50W. It really isn't worth any time or money fishing for something just barely perceivable. One of the biggest ripoffs in audio is charging consumers extra money for imperceivable gains in power by listing it as linear numbers instead of logarithmic. If anything the maker of that amp saved you money by not making you buy a power supply you don't need and instead giving you longevity by making sure the amp has plenty of headroom in thermal rating.
The purpose of my test was to show what you really get from it. The amp wasn't advertised as 35 watts into 4 ohms with one channel driven, it was for 50 watts per channel into 4 ohms with both channels driven. The amp is probably capable of that, but not with a supply that rated for less than 100 watts.
Looks like a breeze audio something I spent $23 for on eBay was surprised at how good it was for such a cheap little thing. Are used a 24 V linear power supply that I built for a Nelson bass amplifier I built it worked much better than the PowerBrick. But I only need one WATT with my 98DB efficient speakers I built one WATT of power is too loud And yes the heat sink inside is way too small for 50 W of any sustained period of time.
That last line is the key to everything. Everyone wants the world for a $1. Realize that what you get, for what you give is all that matters.
So... John is into electronics also... how cool is that!
Merry Christmas and happy 2021 to you and your family. Take care and be happy !
Thanks! Same to you!
I have been watching your videos for few years and im very happy to see that you gotten into audio stuff, because i recently acquired same interest. Got myself amplifier and thinking about building speaker boxes for varies speakers i have laying around.
These Class T amps are impressively efficient. I've used one in my shop for years and it hardly heats up, which was great when I ran it on a solar-charged battery. Nice to see some real numbers from you on this, thanks!
I have the exact same amplifier and I was running into problems with clipping and also just an overall unpleasant sound using an inadequate power supply.
I had a cheap 24v 20A switch mode power supply lying around (the type used for security cameras) so I hooked that up and was able to drive the amp much higher without the clipping and a significant improvement in the sound quality.
The power supply only cost around $20 and so far it seems to work really well.
Great video. Even better shirt !
Great content, well done John
Thanks for this video! I really wanted to know how they were!
Thanks - good to know. It'd be useful to rerun the same test at 200Hz through to 15kHz to see how frequency varies the clipping point.
I concur. I used a 36v power supply and had far better output results. This for some reason also resulted slightly cleaner audio.
I have a similar amp (if not the same), with a 32 volts powersupply and it can play a lot louder than I need to.
I am so satisfied 👌🙏
Okay, so I checked again. I actually have to plug it in to the amp first and then the wall after... Otherwise I will burn it in an instant. For now it works just fine with my pair of Tapco 6915 speakers. Does not handle the bass as well as my NAD, but it sounds really clean. Guess I just been Lucky 🙏👍
Thanks!
(I saw just now, The amp I have has the TDA 7498E chip.)
Could you please add a link to your benchtop power supply build video that you mention? I can't find it, but I need a really good benchtop power supply, and I can't afford to buy a new one that puts out what I want, so I'd be curious to see how you made yours!
Great video .. Thanks man .. 🙂👍
I feel like I saw this video before, a couple of years ago.
I've got a few of these chinesium 3116 based amps at home. I find that a UL listed laptop power supply generally hits the sweet spot. 19V might be pushing it for some of these, you have to check the caps.
I have a hifi and the curse of the golden ears but for 18 bucks plus a trip to the thrift its good enough for my workshop.
If you look more closely at the specs you'll see those power ratings are at 10% distortion. They make about 2/3 that max power with 1% distortion.
There's another thing to consider with measuring this way, the frequency range of the meter. Most meters of that type have a rather low frequency limit for AC volts, meant for measuring household current. While difficult to get precision, using the scope is perhaps better. That's not to say a given meter won't measure 1kHz but you need to check the specs.
Measure the input current while doing such a test.
10% distortion might not be all bad considering when listening to music that output is normally momentary. It's a peaks, not over long notes, generally.
Distortion is the main thing with class D. Since the sound is *recreated* at output filtering a not processed at all in the linear region of the amplifying device there is virtually no noise. And specs say so. The signal to noise ratio is outstanding for such an amplifier.
Yeah, bought one of those chinese S.M.S.L-amps, looking very similar to the one you got there, rated at 2x50 Watts into 4 ohms, with a 24 VDC supply. What they ship it with however is a 12 VDC brick. Not a word about that in the description. Contacted them and they had some poor excuses but promised to send a better power supply. Some time later it actually showed up. 18 V. Hahaha. They didn't lie. It is a little better...
I have a similar amp based on the same chip. It measured at 32db gain, which is way too high.I lowered it to 20db, which resulted in a more linear volume control function and far less noise. I was actually quite surprised that it could drive my B&W Matrix 805s to decent levels.
You'll find that rating is for at 24v and they got it from the chip spec sheet. But they include an 18v power supply with too little of current supply. Sometimes you might find a 24v power supply included, but even that is often no good too since the rated current output is too low to provide the wattage the board is asking for at that voltage. You can get the right power supply for about £20 here in the UK, probably cheaper over in America. The way you calculate if you have enough is by getting the supply voltage and current that the supply is rated for, multiplying them and dividing them by the amps efficiency (e.g. /0.8). That tells you if it's capable of providing enough for your amp. In this case 24v is the max for the board, so get a higher current output capable one to get to that figure you need!
very cool video. i'm into the "audio" world but i don't know much. you should do a similar video with that 21V 8A PSU that you mentioned!
I have work with these class D amplifier from TI, they have one chip that is rated at 100W/ch with almost no heatsink! In continous you can not expected more than 10W/ch without a thermal cutout! Plus these chineese amp PS are underrated for the expected power. Maximum power is rated with distortion at 10%! You can't add any big heatsink...
Those TI based amplifiers are pretty nice (and cheap!). There are definitely some amps that cut corners, but a lot of them look pretty good. 50c says that a 19.5v is a common laptop power supply to keep things cheap, and that sourcing an acceptable 24-26v PSU would bump the cost a bit much.
You don't want to be running higher than 24v on these boards, supplies rated for a higher current at 24v is perfectly fine though
Doin' My Part Brother👍 I think I would have liked to see the weak link power supply theory put to the test. (not at all mocking you, I just would like to see the difference)
You should've shown how/if it's power output improved when powering it with your big homebuilt benchtop power supply you showed there instead of that tiny brick power supply, as I'm sure your benchtop supply could actually deliver what that amp really wants in order to perform at its best.
Try to also measure the output of the power supply under load. Was the amp taxing the power supply? Plus, It would be interesting to see the efficiency of the amplifier.
Did they actually originally rate it at 50 W RMS, “Program Watts” etc.
21V peak is ~15V rms or so with a bit of dropout just about enough for 50W in 4W, it is class-d so heatsinking shouldn't be a big issue
Thing is, it's rated for 50v RMS not peak. The issue is the power supply 👍
When measuring AC power, the power is measured differently as we have a periodic voltage and current. So the true rms power is much less than what you measured as the equations apply for continuous dc voltage and current.
It should be something like P=(1/2)*V*I
He measured the RMS voltage with the meter. His calculations are fine.
He measured AC volts in Vrms. He used the correct formula for power (Vrms^2 / R). I think you are somehow a bit mixed up about something. I am an electrical engineer, so I have studied this stuff and use it all the time. I am not mixed up.
@@mckenziekeith7434 I am not sure what I am missing, also I am not quite an engineer yet. But 3rd year electronic engineering in university.
@@kaumohlamonyane272 you are missing the significance of "RMS." RMS voltage is DC equivalent when it comes to power calculations.
I have a question. Back 25 years ago when I was ripping out my back seat and filling that area and hatchback with speakers, I remember them offering 4 ohm or 8 ohm on the speakers I purchased. My question to you is if it theoretically splits the power in half, why would you want to go with 8 ohm speakers over 4? Or is it you want to match the speakers to the amp? Like I said before, I crammed 4 12's and 3 15's in the back of an eclipse hatchback, so sound quality wasn't as important as shaking grandma's fine China off of her display shelves...
I learned alot over the years in the field I was playing in. I was in what was known as decibel drags. Basically pull up your car, they stick a decibel meter in and you see how high you can register. I learned how far the sound waves need to linda bounce around in the car before hitting their peak, the angles the speakers needed to be to obtain that and temperature of the air in the car for best results. But as I stated before, quality of the sound wasn't extremely important, just more loud the better.
An 8 ohm load is easier for an amp to drive, especially a low cost one. An amp will try to supply double the current into a 4 ohm speaker but it needs a larger power supply (transformer, usually) to do that.
@@IBuildIt interesting. I was always curious and watching your video today kid of put it into terms I understood. Thanks
@Douglas Blake is there a specific use for 4 ohm? I mean it sounds like everyone would just want to use 8 ohm.
sometimes I believe a rating if the amp is encased in a heatsink.
This is why a single slot car stereo only gets max 40W per channel.
To get more, an internal power supply to step up the voltage is required, but all this does not fit into a normal slot, let alone the necessary heat sinks.
You had power supply on the bench right there. Revisit please.
My homemade supply has the voltage, but I'm not sure it has enough current.
Can you test the power supply to see how much it's max output is? That would be useful to know.
Also, you need to look at the spec sheet for the amp and see what the specs really say, on good quality amps it will say "50w @ 4 Ω at .01% THD" but on cheap amps, it will say "50w @ 4 Ω at 10% THD" which means essentially a square wave and 1000 times more distortion than the first amp. You have to really read the specs and hope they aren't lying. Some cheap amps will just outright lie on the spec sheet.
Crank that thing up till the speakers catch on fire and everybodys UHF tvs go fuzzy
What is a UHF TV?
Build an amp with a flux capacitor. Love the new channel!
I used to laugh at those flea market amps that had 300 Watts on the box. You just know they are measuring peak to peak instead of RMS (.7 x p-p/2) values. Then try pointing it out the the vendor who wouldn't know a volt from an ohm.
Resistive loads are not the same as inductive loads, I bet it delivers even less power through a speaker coil
speaker is complex load involving resistance, capacitance and inductance; dummy load just give us a "close enough" idea of raw power
Electronics. Course on makers mob? It's common, when u buy an amp ... That the weak point is the power supply, so the amp doesn't have as much output as it could?
Beware running any amp continuously at its nominal power rating - even 'good' ones will overstate what they can really survive continuously and I have popped an amp doing exactly this kind of test.
Actually not the case with this board. I've tested it full load 24v (with a heatsink of course, that's needed at 24v) for an entire 24hrs with no problems
1,21 Gwatts
You're not going to hear much difference in loudness between 35 and 50W. To sound "twice as loud" you need 10x the power. Most people can start to just perceive a difference in loudness at 1.5 dB(SPL). That's 1.4x the increase in power. And that's precisely the increase between 35 and 50W. It really isn't worth any time or money fishing for something just barely perceivable.
One of the biggest ripoffs in audio is charging consumers extra money for imperceivable gains in power by listing it as linear numbers instead of logarithmic.
If anything the maker of that amp saved you money by not making you buy a power supply you don't need and instead giving you longevity by making sure the amp has plenty of headroom in thermal rating.
The purpose of my test was to show what you really get from it. The amp wasn't advertised as 35 watts into 4 ohms with one channel driven, it was for 50 watts per channel into 4 ohms with both channels driven. The amp is probably capable of that, but not with a supply that rated for less than 100 watts.
Sixth.
I have a shirt that says"No one cares if you aren't a decent human being.".😉
You get what you pay for.
Looks like a breeze audio something I spent $23 for on eBay was surprised at how good it was for such a cheap little thing.
Are used a 24 V linear power supply that I built for a Nelson bass amplifier I built it worked much better than the PowerBrick.
But I only need one WATT with my 98DB efficient speakers I built one WATT of power is too loud
And yes the heat sink inside is way too small for 50 W of any sustained period of time.