Sure Electronics -- New low cost 25w-100W board

[sled108]

Parts Express
Sure 2x100W @ 4 Ohm TK2050 Class-D Audio Amplifier Board

[trevmar]

Ah yes - silly me

The inductors SURE use on that board (7B12NA eqiv) saturate at around 3.2Amp, and that is not 100W, it is (P=Isquared x R) more like 50W into 4 ohm. So I am not sure I would place 100% confidence in the rest of the specifications of that board, either.

I am 99% happy with my HifimeDIY T2 (the -1% is because it is larger than I would like) and I was looking to SURE for a really small amp. I am actually still using the SURE TDA7498 board solely because of its size and weight advantage, but I would tend to look at HifimeDIY before I looked at a SURE board again... If I had a split power supply I would go for the L20D without hesitation...

I was surprised that after I had tweaked the TDA7498 board, changing the CPU and the other components, it sounds a lot better than when it arrived from the factory. Back then, I took away the MiniDSP 9Khz +2dB preemphasis, as the original board was 'quite bright' (because of the high level of HF distortion, I guess) but I put it back after the board had been 'brought up to spec' by my changes. I shouldn't have been surprised, but I was. Psycho-acoustics never cease to amaze me...
.

[sled108]

I've run my TK2050 into 4 Speakers with 32v...It puts out a Very Impressive 50 Watts..

[Saturnus]

Core saturation isn't a major issue since it would only occur at peak outputs which are distorted anyways. With normal music it's not a problem.

And yes, the TK2050 amp is far better. Given my experience, I'd not even regard the TDA7498 good enough to drive a clock radio. Might be ok for a sub though.

[trevmar]

.
I am gradually getting some decent performance from the TDA7948L I put on the SURE board. The original chip on it was probably OK too, but I know the new one is from a full-spec source.

When SURE designed this new board they decided to use surface mount capacitors as much as they could. In particular, the capacitors in the output network were X7R ceramic. When I changed them to Metal Film I finally started to get some half-decent performance out of this chip.

Attachments 1 and 2 are the photos of the board top and bottom, showing how I fitted the film caps onto it. I designed the new output filter to be critically damped at 8 ohms, which indeed it turned out to be. The main filter caps, C18 and C14 on Sure's AA-AB32189_Ver1.0 schematic, are 330nF film. The four caps C15, C16, C19 and C20 are 100nF and the four snubber caps I added are 220nF in series with 10ohm 2W resistors. The 100nF caps are soldered on the rear of the board.

I compromized between having a high attenuation of the Class D switching frequency and having a good clean audio response. You will see significant high frequency signal on the output, but it is not nearly enough to zap a tweeter.

The 1KHz square wave responses with 6 ohm and 12 ohm load are shown in the scope photos of attachments 3 and 4. One is underdamped, one is overdamped, as the design point is critical damping with an 8 ohm load. Attachment 5 is the frequency response with an 8 ohm load, showing a correct, critically damped, frequency response in the 10KHz-20 KHz region.

Attachment 6 is the THD vs Frequency curve that used to be so bad. The distortion is now acceptable, staying below 1% even at the highest frequencies.

To double-check this I took the distortion vs power curves at two different frequencies, 1KHz and 6KHz, just to make sure that the HF anomalies are now under control. The curves are in attachments 7 and 8. I used a 6 ohm load, the maximum rated load on this amplifier (although mine does drive 4 resistive-ohms quite happily). Note that I have spilt out the even order harmonic distortion components in blue, and the odd-order in red, as is my habit when trying to match up how an amp sounds, versus how it measures

So there we have it.
IMO SURE have managed to blow their reputation for reasonable audio boards by putting out a poorly designed TDA7498 board to replace their previous TK2050 staple. The decision to put X7R caps was probably reasonable enough, they save a dollar or two, but they should have detected problems when they started acceptance testing, with the same debug process I just had to go through.

This board is selling on Ebay right now for less than $25 delivered. The TK2050 board is still twice that price. I guess I am not the only person who has been disappointed by the new board's performance I am, however, now enjoying using this 'modified', 'fixed', board with my micro-weight suitcase-portable system. It is not audiophile, but close enough, at least until something smaller and lighter comes along...
.

[Juicie]

Have you contacted Sure about the issues with the board. What do they say about it?

[trevmar]

Originally Posted by Juicie

Have you contacted Sure about the issues with the board. What do they say about it?

Juicie,
I travel to China a lot, and spent some time in Shenzhen earlier this year, wandering around the electronics markets. My impression is that the nascent surge in electronics design in China is just like I saw in the 1980's here on the US West Coast. Individuals building great prototypes, and entrepreneurs springing up to help them distribute and commercialize their work.

So you see Nick at HifimeDIY helping us to get access to the wonderful Chinese amplifier modules he ships, and you see the Hong Kong retailers marketing the L20D modules designed by ljm_ljm. If you go to Taobao (the Chinese version of Ebay) you can find that these designers primarily market their product directly to the public.

For example, here is the Google translation of ljm_ljm's Taobao storefront: Google Translate

So a Chinese DIY amplifier company does not seem to be structured along the same integrated model as we now see with the larger companies here in the West. As far as I can see, SURE has no resident technical expertise. They are marketeers. When I recently tried to get a spare potentiometer for the TA2024 based amplifier I had bought a few months ago, there were numerous exchanges with technical support, and finally a knob arrived in the mail. Not a potentiometer, but a knob. They were trying to help me, as they did did send the knob for free, but failed to grasp what a potentiometer really is. This is despite us having agreed on my photograph below as showing the problem component.

So I haven't tried to communicate these technical issues to the marketing and support staff at SURE, I am convinced that would be a waste of my time. I am hoping that the designer of this board will find out about the information here, and that I can share with him my experience getting this board to function acceptably. It does sound OK now...

If anybody has direct contact with the board designer, please send me a private message with the contact details...
.

[lithoc]

Originally Posted by trevmar

When I recently tried to get a spare potentiometer for the TA2024 based amplifier I had bought a few months ago, there were numerous exchanges with technical support, and finally a knob arrived in the mail. Not a potentiometer, but a knob.

IMHO, language barrier is the main problem.
I goto taobao often, it's all mandarin.

When someone copied their product and paste it on aliexpress, they use google translate without checking them.

The translation goes as far as google translate can do.

In mandarin, a simple 1 word can have multiple meaning.

[larzman]

Originally Posted by trevmar

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

PE = Parts Express, parts-express.com

[Dr3Tri]

Originally Posted by trevmar

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).
.

Hi, I am beginner so bare with this stupid question of mine.

I know that tripath amplifiers do not consume a lot of electricity, but I have seen that ppl. recommend big and expensive PSU's for those. Hows so?

If this 2x100W amps consume just few watts, couldn't I run those with tiny little PSU too?

I bought 4x100W tripath amp from Sure Electronics couple years ago and tried to measure how much it draws energy, with typical common energy meter. It shows so little watt amounts that I tought my meter is either broken or this kind of amps can not be tested with that mater at all. So that those readings lie somehow.

I just checked again, I am listening music with comfortable level and meter shows it takes 7W. Could this be correct?

Which kind of power is recommended for this 4x100W amp? At the moment I use it with old ATX PC's PSU (I don't remember how powerful it was, I took it's electronics out of the case and inserted it to old and broken Marantz amp with the Tripath amp.
I could use Marantz's power switch, RCA inputs and speaker outputs. So I don't have power ratings for that PSU anymore.

Now I tried how much meter shows when I turn up the volume. I put music )(and bass) so loud that I could without sound going totally distorted and it took only 1W more, 8 watts.
So it eats basically same as my DJ mixer (7W) which is in front of it.

If this amps use so little power, I guess that some car battery would work well for a long time..?
I've read that batteries give better sound quality than AC to DC converting PSU's. Is it true? I plan to use 70Ah battery with some tripath amp, not necessarily with this 4x100W though...