diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Class D (http://www.diyaudio.com/forums/class-d/)
-   -   Sure Electronics -- New low cost 25w-100W board (http://www.diyaudio.com/forums/class-d/194413-sure-electronics-new-low-cost-25w-100w-board.html)

trevmar 11th August 2011 05:40 AM

Sure Electronics -- New low cost 25w-100W board
 
8 Attachment(s)
I just bought one of the new SURE Electronics "2 X 25Watt Class D Audio Amplifier Board - TDA7492." When it arrived it actually had the TDA7498 chip on it, with the 2 X 100W capability (no fan on the heatsink though, and you would probably need one at 200W :) ). I paid $23.66 for it - a bargain.

So, even though I might be accused of 'looking a gift horse in the mouth', here are the problems which I found in this new SURE design. There are 1500uF of supply capacitors, made up of seven 220uF/50V caps, which is probably enough, although barely. The board is well laid out, with a gain switch for 4 selectable gains. Ground and power planes are designed well. The heatsink is easy to remove without damage - don't lose the spacers, though...

However, SURE need to send their designers to a training course on Class-D output filters. The data sheet for the TDA7492 shows, for 8 ohm, 33uH and 220nF, with 100nF to ground and no Zobel output damping network (to understand why these components are there, look at, eg: Class-D Amplifier Output Filter, Class-D Output Filter, Class-D Amplifier, Audio Power Amplifier, Low-Pass Filter, Common-Mode Filter, Sinle-Ended Output, Yun-tao Zhao, Class-D, Audio Amplifier, BTL Output, LC Filter, Hybrid Filter, John Widder, Audi

For the TDA7498 the suggested inductor has changed to 22uH, the main cap is 470nF (for 8 ohm), with 220nF to ground. A Zobel damping network of 200nf and 8 ohms is used with the more powerful chip.

When SURE decided to change to the new, powerful, TDA7498 chip they clearly didn't bother to change their output circuit. Actually, what they have on the board wasn't even correct for a TDA7492, but I have already suggested remedial training for these designers :) . Anyway, as you might expect, the board is unstable.

Attachment 1 is a sinewave driven into 8 ohms dummy load showing oscillation of the Class-D negative feedback loop, and Attachment 2 shows how it gets worse as the amplifier is driven more towards its rated power. The square wave response is in Attachment 3. Dismal indeed.

After disassembling the capacitors C14, C15 and C16, which can be fairly easily accessed (see attachments 4 and 5) I found that a 1.0uF was used for C14 and 0.1 for C15 and C16. Since SURE used a 33uH inductor rather than the 22uH suggested for the TDA7498, the output cap, C14, should have been even smaller than the suggested 470nF at 8ohm, or 680nf at 6 ohm. In my case, that means around 470nf for 6 ohms. I didn't have any 470nf surface mount caps which would handle 50V, so I used two 220nf in parallel (by the way SURE, what is the voltage rating on the capacitors you are using?). I used two 100nF in parallel to get to 200nF (approx) in the C15 and C16 positions. I used poly 220nF and a 10ohm 2 watt resistor for the Zobel damping circuits. The fixed assembly can be seen in Attachment 6.

Now the amp is working well. Attachment 7 shows the new 440Hz square wave response with a 12 ohm load (a light loading, worst case) and Attachment 8 shows the square wave into 6 ohms. The sine wave output is now stable, of course. There is some fuzzyness on the peaks, somewhat similar to that in the Tripath Class-D drivers, but a little different. It doesn't affect the distortion or sound, however.

The photos only show 1 channel. The second channels has to be changed in the same way as the first.

So now I have a nice little amp, with a sweet sound and a big punch (on a 24V Meanwell I was getting a good 50 Watts). The weight is only 6.1 ozs, which is why I bought the amp in the first place, a tiny amp for my suitcase-travel kit.. (more about that in a future post).
.

trevmar 11th August 2011 01:59 PM

Having had the chance to listen to this SURE amp for a short while now, in addition to measurements on the test bench, I am pretty impressed with this 'new' chip -TDA7498.

The Undervoltage works superbly - there is no click in the speakers either when you plug in the power, or pull out the power cord. That is impressive.

I can't hear any residual noise when I put my ear to the speakers. The same settings give 40W max, which would deafen me... Actually, there is a faint low frequency pink hiss, if I align the speaker perfectly with my ear canal... Maybe that is MiniDSP noise, I am using the (digital) volume control connection on the MiniDSP as the master gain control

I connected the SURE to the same low-distortion Peerless 830983 speakers I have been using with my HifimeDIY T2. Both systems can easily show the quality difference between FLAC and 160K MP3 (a subjective test which quickly evaluates clarity). Without the filter modifications (above) there was a noticeable treble emphasis with the SURE board, but that has mostly gone away with the modifications. SURE above 10KHz is not as flat or clean as the HifimeDIY T2, though. A slight 1dB peak at 16KHz was measurable on the bench, in addition to the subjective observations. You can see it on the square wave (above) and I confirmed it on the 'scope with a sinusoid sweep. I will run a distortion vs frequency test in the next day or two and see what that tells us.

If size, weight and cost were not factors, I would still choose the HifimeDIY T2 for subjective and objective audio quality. It is just a better thought-out device. But this new SURE board fills a niche for smaller, lighter, cheaper, more efficient systems (the T2 draws around 6 watts with no signal, the SURE around 1.5).
.

Randallm 15th August 2011 07:35 AM

I bid on one of these boards on ebay and 'won' it before reading your post. I haven't paid for it yet, and I'm wondering if I should cancel the purchase. I'd be willing to make the same modifications when the board arrives, but my skills are pretty rudimentary. I'm wondering how difficult it is to remove the surface mount components and replace them. The picture is a little fuzzy for me. Are you willing to guide a newbie through the mods?

First, how does one remove the surface mount caps? Do you just heat their connections with a soldering iron and pluck them off the board?

Second, what do you do after removing the original C14, C15 and C16 suface mount capacitors? It looks like you've connected the new capcitors to the resistors, and to each other. But where are they connected to the board? Are you using the connections to the board at C14, C15 and C16, or are those components just removed?

trevmar 15th August 2011 07:44 AM

My suggestion would be not to try soldering or desoldering surface mount components if you haven't used them before. I am currently retesting the distortion figures I posted for the board, and should be reposting them again tomorrow. Maybe I made a mistake the first time around... but if not, then the board not meeting the chip's distortion spec would seem a good reason to cancel the transaction.

That said, I am keeping my board, as I need a small, lightweight 25+25 amplifier for my travel kit. Which is why I am spending so much time getting it working properly.

At some point SURE may bring out a revised PCB for the new TDA7498 chips - you might email and ask them whether they are considering that.
.

trevmar 16th August 2011 03:46 AM

I have retested this amplifier board I received from SURE Electronics which was labelled as: "2 X 25Watt Class D Audio Amplifier Board - TDA7492" and which came with a chip whose top was laser-embossed as 'TDA7498'.

This amplifier seems to be using a chip which does not conform to the distortion specifications in the data sheet for either TDA7492 or the newer TDA7498. My best guess is that the CPU with the board is a "seconds" chip, which, eg, 'fell off the back of a truck'. I have tested the TDA7920B chips from ST microelectronics, which largely perform to their data sheet. But the SURE amplifier has harmonic distortion which is unacceptable, and way above the ST data sheet values of either chip. I cannot believe that SURE could mess up their implementation so completely, although that was my first reaction, and have come to conclude it must be a second-rate chip.

Please find the arguments in support of my conclusion in the other thread at post: http://www.diyaudio.com/forums/class...ml#post2673643
.

Randallm 16th August 2011 08:50 AM

I've often wondered if they're using 2nds. I'm not surprised to find that they are (though I am surprised/impressed that you figured it out). I had already cancelled my order based on your advice. Thanks.

trevmar 16th August 2011 10:54 AM

Rnadallm, I have ordered a TDA7498 from Mouser Electronics, and when it arrives I will put it into the circuit and do the testing again. I did notice that ST produces several versions of this chip with different specifications, and even a mono version with only one amplifier active. So there may be some form of selection going on after manufacture which might allow seconds to reach the market, or SURE may have really broken their circuit board implementation somehow. I am not sure. But I went over every component on the board, and except for the inductors (which I replaced) and the 330pF/22R Zobels (which I added) nothing jumped out at me as being different from the data sheet recommendations. I will have more data in a dew days. I do really want to get my board working well, as it only weighs 6.1 ounces :) A good match for my MiniDSP :)

trevmar 20th August 2011 05:40 PM

5 Attachment(s)
I bought a TDA7498L from Mouser. When it arrived it had very different markings from the Chinese chip, but, honestly, if it performs any better, then there is only a marginal difference. I need to apologize for suggesting (above) that SURE's chip supplies may not have been genuine.

I was very disappointed to find that there was no way I could achieve the levels of distortion printed in the chip data sheets. I modified the SURE board by adding the components from the data sheet schematic that SURE had failed to provide - 1nf caps from each input to ground, and the 22ohm/330pF snubbers on the outputs. I also changed the output inductors to 1D14A-220M high-current types (also from Mouser) without getting close to the data sheet performance values (so I changed back to the original inductors). Recall that I have already (see above) changed the output cap values to match the data sheet values. Especially troublesome was the performance at high frequencies.

Having said that, the amp doesn't sound too bad at all (now) and I will in fact use it as my suitcase-portable amplifier. The SURE is not 'audiophile' quality however, I will use my HifimeDIY T2 when weight is not an issue.

So here are the final test results for the TDA7498L. First, I have attached a photo of the power chip on the board showing the additional components I added (white arrows). It is always amazing to me that this tiny, tiny, chip can put out so much power, with not too bad a fidelity...

Then is the curve of harmonic distortion vs power level at 32 volts supply into an 8 ohm load. Apart from its remarkable irregularity (lack of smoothness) the data sheet shows that the distortion I am measuring is 5-10 times what I should be getting.

The distortion levels of 0.5% to 1% are significant, as the best loudspeakers are in the region of 0.1% to 1%, and an amp should be able to do much better than that, so that the amp doesn't become an audible factor.

I then ran the curves for just the second and third-order harmonic distortions, and it is evident that 2nd-order distortion is dominant. Many people say that the 'warm' sound of tubes is partly due to the generation of even, rather than odd-order, distortions, and maybe a similar reason accounts for why this amp doesn't sound all that bad when you don't listen too closely... The even order curve is in blue.

Next is Intermodulation Distortion (IM) at 1KHz and 1 Watt (8ohm, 32V supply). I have certainly seen much worse IM than this, and I have of course seen much better :) It is interesting that the 6KHz IM (not shown) test gives about the same value as the 1KHz IM.

Where the amplifier primarily fails is at distortion higher frequencies. Some would argue you can't hear the second and third harmonics of 6KHz, and I am not going to buy into arguing that point here. I do note that the IM distortion does not rise with frequency, which is a mitigating factor for the harmonic distortion. The final curve is distortion vs frequency under conditions as nearly the same as I could get to the data sheet curve - 1W into 8 ohm, 32V, 39K, etc. Yet the measured curve is >10dB higher than the data sheet values...

So there is is. The TDA7498L chip data sheet shows that this out to be a much better amp than I could replicate, at least using the SURE board. As long as you don't try to run it near full power it works reasonably well, but it is not 'audiophile' quality... and please don't expect it to sound nearly as good if you stress it towards the advertised 100W output levels.

Unless you have significant experience soldering and desoldering surface mount components, this is not a board you should think of modding. The TDA7498 is a tiny chip, with a lead pitch requiring a steady hand, and the correct tools, to handle properly.
.

sled108 23rd August 2011 12:26 PM

Sounds to me that the TK2050 from PE is still a Better Deal....

trevmar 23rd August 2011 12:38 PM

"PE"? I'm a newbie here - who is PE?


All times are GMT. The time now is 05:42 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2