Omnitronic EDP 1000 Watt Class D Amplifier Teardown
Vložit
- čas přidán 27. 07. 2024
- gerrysweeney.com/omnitronic-ed...
In this video I describe the basic operation and topology of a Class D amplifier and then tear down an Omnitronic 1000W amplifier to have a look and see what is inside. - Věda a technologie
It's amazing how small amps have become for their power. I used to work at a music shop in the early '90s and we had a used 500w+500w class AB amp from the late '80s for sale and that thing was an absolute beast, it was only 2U high but nearly 3 feet deep. Had two massive toroidal transformers and rammed with MOSFETs and needed two people to lift it. I don't think anyone ever brought it in the end 😆
I think the reason you see poor performance at lower output volumes would be due to limitations in how precisely you could time the pulses. At 10% of output voltage the DC would be varying between 45 and 55%, meaning a precision of 0.01% minimum would be needed to have 60dB dynamic range. I think increasing the oscillator frequency would help in that case.
Obviously any jitter in the timing or hysteresis in the comparator + frequency modulation of the oscillator would generate output noise.
Glad to be of help, thank you for watching
Thank you for the feedback and the subscription. Gerry
I fix Audio Research audio equipment for many years, fix amplifiers is an art.
Cheers.
I think the relays are classic output protection relays, i.e. they cut out when excessive clipping is detected. I say that because there is a third speakon socket on the back which you plug your speaker into for bridge mode, all that needs is to be wired across the L+R hots so no switching is required. Thanks for the comment. Gerry
Nice video. The relays in the output stage are likely to be for switching the two channels onto one speaker connector for use in bridge mode.
Wonderful video, subscribed. Can't wait for more similar vids. Thanks for the upload.
Hi Gerry,If you notice a 'spike' in Hits from Ireland - We shared your excellent Video on our Facebook - Dicker Event Tech.
Thanks for a great explanation of Class AB Vs. Class D.
There's more to being a Sound Engineer than 'Pro Tools' & 'Plug-Ins'
Good luck with the project, building one would be interesting for sure. Gerry
Hi stuckinpants, thats for your comments, yes I agree with you. The problem always seem to be in designing a good output filter. As soon as the frequency goes up the complexity of filter design and the modulation, radiated RF power and other factors start to kick in. It seems that most average Class D designs are a compromise and end up powering sub's and PA's and car stereo systems. Not a bad thing really. Gerry
Class-D Audio technology, Good feature for the future!
That switch is needed! When you throw it into "mono / bridged mode" it causes the input to both amps to be 180 degrees out of phase. That's how you get such a huge power increase!
Hi,
Its now in use in a school and I am not sure I can remember 100% but I am pretty sire that air is drawn in through the front and blown out of the back. Next time I see it I will check to make sure.
Gerry
Very Very Good video, hope you have another coming up on repairing boards.
I have a class d amp which in high mid frequencies skips one of the signal LEDs and goes directly to the clip .. or else they turn on both together .. with gave frequencies if you mark the signal LEDs correctly until entering the clip
Impressive indeed I love the knowledge
Nicely Eleborated 👍👍👍
Excellent explanation! Thank you! :)
very nice! got me quite a lot better understanding of class D amps :)!
Yeah it was a rush job I am afraid, it was for our local school, I done a quick teardown before I put it into use. Its pretty powerful and pretty decent quality sound for a Disco PA at least.
where did you buy it so cheap ?
thanks for the video. Very clear and useful !
a verry good video garry
@TheBinary yes you are probably right, all I was saying was that in my experience thats where I tend to see them being used. I know they can be very good, I just don;t see that many of them, they still don't make the mainstream which to me is a good indicator that its difficult and expensive to make good.
Nice video.
Too bad that you didn't power it on so we can hear quality and you could conect it to scope so we can see diference beatween input and output.
Looking at the ETD39 core I don't think this amplifier will go to a 1000W. The power that this PSU can supply would be in the 400W range. Dynamic power could be higher of course. The power supply has universal input range, has it got active PCF?
Gerry,
Is the airflow on these rear to front? Looking at where the fans and heatsinks are, I am guessing so, but I'd like to confirm this. I am asking as it needs to match the rest of my amp rack (2xiNuke 3000DSPs) to avoid getting hot air recirculated around my amp rack and possibly leading to early thermal failure :(
Thanks :)
@TheBinary, I did guess that you like Class D amps ;) Always liked the look of the Hypex modules. I am going to try and do some testing of a Class A/B amp vs. a Tube amp soon, I should probably test a Class D amp along side at some point. The Hypex modules always looked like nice implementations, I like the blue T-shaped heat sink. Gerry
Gerry Sweeney
Thanks
for 2kw power you need minimum 72v per rail into 1.3R
How much price & how can take it in oman..??
@TheBinary, my goal is to explain the principles I understand which may or may not be correct (see my blog about page) - I will leave that up to the viewer. MOSFET's are more like voltage amplifiers in practical terms because of the way they work, at least thats my understanding and its born out in practical circuits. The point I was making about the filters is they are hard to design, and yes I do see HF on speaker outputs of that amp with my 200Mhz scope so the filters are by no means perfect.
The schematic you used to explain the Class B amplifier is demonstrably wrong.
Sorry Gerry, correct me if I'm wrong, but the schematic at 3:00 on the inverting input of the op amp you have that voltage divider set up kind of weird.
the way that the divider is set up basically makes that top resister completely useless (assuming ~0 Current flow). I assume that you want that divider to step down the output voltage. just thought to point that out.
great video though, big ups on the awesome explanations!
+Erick Miller yes you are right, I was just doing a block diagram and should have drawn the resistor from the output and tapped the input to the op amp from the centre point, I should have drawn that better, thanks for pointing it out. None the less I would not recommend anyone try to build what I drew, as I say it was merely a block diagram :)
You saved me making the same comment.
Nice channel there! Suscribed! ;)
I would guess, it is possible to do the analog PWM generation with a DSP instead?
Great vid.. thanks
very nice thanks
nicely done .. !!!
ElectronicBob thank you. Gerry
hi Gerry im thinking of buying an omnitronic 31xl audio analyser just wondering what you thought of omnitronic's build quality not really familiar with the brand
Hi Johnny, I don't know the analyzer but going by the build quality of the class D amp alone, they are pretty good. I though the thing was well constructed and well designed. Gerry
thankyou Gerry for replying regards
Hi, I guess so, thanks for the sub.
Not sawtooth, triangle pwm! Cause you need to symmetrically compare signals, not rising edges slower, falling edges faster (as would sawtooth pwm do)
Many thanks :)
What you have shown on the drawing is a typical B class amplifier where one transistor works for positive half of the signal and the other one works for the negative half. While beign more efficient than A class amplifiers where one transistors handle both positive and negative signal parts, it's disadvantage is that the transistors do not switch instantly so you have crossover distortions in result.
Class AB solves this problem by feeding a little bit more than half of the signal to both transistors.
+tuomollo class AB amplifiers work in class A where small signals are involved, the actual mode is a double-ended class A amplifier where both transistors are biased on at the same time. Only when the signal swings in one direction or another does it become pure class B. I did not show the bias configuration so the illustrative diagram could be either pure class B or class AB, I think in the video I was talking about class AB.
+Gerry Sweeney I was reffering to small sine waves you were drawing while explaining. You did simplify things a little but it's OK
+tuomollo yeah I was trying to keep things simple and showing the difference between a linear amp and a switching amp. An explanation on A, AB, C operation would be another video :) Thanks for the comments
looks like a eevblog lab (the place where you are)
Its my own lab, but yes I know what you mean, most electronic labs have similar characteristics...:)
Gerry Sweeney
yep
good
what is best for car audio ? ab or d
Either works well, class D would be more compact and dissipate less heat but are probably more expensive. In both cases the quality and efficiency of the PSU which in the case of car audio would be a switch mode, is predominantly what dictates the power capability. Gerry
no joke, i bought a D-CLASS precision power p900,4 $1900TTD
ouch!
tell about it
Its hard to believe that amp would give 1000w rms...
exoticg9 thats the great thing about class D, think of it like a switchmode PSU in terms of efficiency, same sort of technology really. Gerry
it´s 1000W bridged on 8 ohms but only 300W per channel on 8 ohms - that´s possible
+Jose Francisco Medeiros Class T is a D class amplifier with a different name.
6n137 opto used as gate switch new tech
As a man dealing with audio I would care about audio quality of my videos...
I do, see my other videos, I was learning as I go :)
115 quid , there is no way that is any where near 1000w rms then , my m8 just paid 1500 quid for a true 5kw PKN amp , those omnitronics are horrid cheap things.
I suspect you could get 1000w out of it - what it would sound like though who knows - I should have benchmarked it but I did not have a suitable load to test it with. I would not equate price with quality though - that does not always corrolate
That is true , but mostly with power amps they lie lie and lie about specs all the time , especially cheaper brands.
Darren Woods We used it for a school PA system and it did not sound too bad at all, but it was just low-cost PA speakers. The previous amp it replaced was four times the size and five times weight was 300W per channel and the Class D amp sounded much better by comparison. I do not disagree about the mis stated specs though, I have seen many a car amplifier claim to be 1000W :)
If I get the chance (and a suitable load) I will try to grab the amp for a couple of hours and benchmark it for power and output distortion - will be interesting to see how close (or not) the Amp gets to its claimed specification. As a more general observations, the electronics and the physical construction were both surprisingly good given the cost of the thing. Gerry
Oh now that I would love to see personally. And yes have seen a lot of shoddy 600-1000w amps on ebay for cars with only 20amp inline fuses.
Also recently I worked on a amp for my m8 that was a real old but a monster 5kw(2450w rms x2 at 4 ohm) it had 2 massive tor's in there with a ton of copper each , one was reading .4ohm on the primary and it was tripping the breaker instantly so I told him call crown and see if you can get it replaced.
They quoted him 300 quid , and hes now bringing it to a local rewinder to have a look at it , I did say to him well use it at half power and just use one transformer for both sides of the amp as a quick solution.
But he bought a new amp from pkn that is well they call it class G and its 2u and has a lot more processing power also included with it , got to love modern switching psus they are amazing at what they do.
Darren Woods The transformer used in the SMPS looks good for about 300w. ETD39 maybe.
I still like AB better.
Glad to be of help, thank you for watching