rear shot.
Banana plug is bad...
gychang
Even though it may work quite nicely and it is built solid like I did in the past years I can see that there can be improvements made to the grounding. (just like with many of the amps I've built in the past years)
Like the RCA sockets can best be isolated from the chassis, then the DC output negative (or ground) from the power supply should be connected to the chassis on one point as close to the bulk caps as possible. This should eliminate some noise and hum.
Then to further fight the hum...for as far as I can tell you have tie-wrapped a mains cable to some of the signal wires? (on a side note: Why don't you turn that power supply 90 degrees clockwise?)
Then your signal wire-routing/length could be reduced to less than a third of what you've been doing now. Nicest way is to place the volume pot and selector on a bracket near the back panel, then extend the shafts towards the front panel.
Oh...there are always so many things one can remark about a design that is already finished.🙁 Just like I always do when I finish one more of my own amps...🙄
It just surprises me a bit with all the experts here on this international forum nobody has even made the slightest remark about this?😱 Aren't we all here to make us all improve on our work?!
Last edited:
I appreciate your input, for this novice some of your comments make vague sense, understand the benefits of shafts...
I know u have pictures of some nice amps you have built that may illustrate some of your points, if not too much trouble are u able to upload your amp picture with some comments?
thanks in advance.
gychang
Hi Gychang,
I want to make a drawing of these points later on when I have some time. Then I shall certainly post it here...🙂
I want to make a drawing of these points later on when I have some time. Then I shall certainly post it here...🙂
long lasting project update
Hello
last december I began to assemble a multichannel amp, including active crossover.
Look here: http://www.diyaudio.com/forums/class-d/112395-sureelectronics-tripath-boards-72.html#post1687960
This project was a long time on ice, because of two problems:
- strong hum in one amp
- very strong power up and switch off noise
Then, I found no time to work on it. Now, after beeing discharged by Studer/Harman and moved to a new appartement/shack, I have the time to work on it.
The hum is gone: one amplifier is grounded in star topology, the other one in bus...
Must be some ground connection from one of the boards. I did not find the difference.
The second problem is solved by a simple delay/mute circuit, based on NE555 and a relay.
In the mute position, the relay feeds 5V (tapped at R1 of the sure board) to the mute connector.
After powering up, the relay remains 5 sec in the mute position, after powering off, the relay immediately switches to mute and as long as the psu caps are feeding the amp, the amp remains muted.
Absolutely silent now!
Franz
Hello
last december I began to assemble a multichannel amp, including active crossover.
Look here: http://www.diyaudio.com/forums/class-d/112395-sureelectronics-tripath-boards-72.html#post1687960
This project was a long time on ice, because of two problems:
- strong hum in one amp
- very strong power up and switch off noise
Then, I found no time to work on it. Now, after beeing discharged by Studer/Harman and moved to a new appartement/shack, I have the time to work on it.
The hum is gone: one amplifier is grounded in star topology, the other one in bus...
Must be some ground connection from one of the boards. I did not find the difference.
The second problem is solved by a simple delay/mute circuit, based on NE555 and a relay.
In the mute position, the relay feeds 5V (tapped at R1 of the sure board) to the mute connector.
After powering up, the relay remains 5 sec in the mute position, after powering off, the relay immediately switches to mute and as long as the psu caps are feeding the amp, the amp remains muted.
Absolutely silent now!
Franz
I found the cause for the problem, finally:
A cold solder point on my stripe board psu...
Now, both amps are working perfectly, internally grounded in star topology.
I feel on the save side now, to attach my speakers to the t-amps.
An externally hosted image should be here but it was not working when we last tested it.
I want to try making a portable amp, sort of like a Tivoli Audio Pal, but more powerful and I have some spare panasonic satelite speakers. They are rated for 100watts, could i power them off of this amp and 15 watts each? 😕
I am looking at the pre-built Sure amp with the AA-AB340 board, can someone tell me if that is an opamp buffer stage on board?
Thanks,
-Steve
Thanks,
-Steve
An externally hosted image should be here but it was not working when we last tested it.
According to the circuit schematic - download here - Download Center---Sure Electronics Co., Ltd. - No!
Tony.
Tony.
I hope you don't mind if I post a pic of my TA2020 based amp?
An externally hosted image should be here but it was not working when we last tested it.
It's not the Sure board, but it's pretty similar... I'm not sure if there's anything I've done wrong here or any improvements I can make? The SMPS is grounded to the chassis thru the screws in the bottom.
This is the first amp I have put in a chassis, and I'm learning as well, but I'm getting no hum/buzz and I'm happy with the performance.
I've used a Molex connector for the speakers, not really the recommended usage, but it works fine. 🙂
Just curious - does anyone know the purpose of the four metalic surfaced holes near the speaker output terminals. (I havn't read through all 88 pages of this post so I'm not sure if this has been previously answered.) I know they're connected to the outputs but what is their purpose?
Some thoughts and experiences.
Recently I bought the little sure board as I was interested to see how it performed compared to my gainclones. I need several amp modules to complete my open baffle system and was about to embark on building my final gainclone modules but felt that digital modules might be a good option as my whole system is "off the grid" and it is much easier to use a single sided supply rather than the dual rail supplies needed for the gainclones.
When I first fired up the 2024 I was not much impressed, it sounded thin, but I knew it takes a bit of burn in so I let it run, within just half an hour it was sounding better and by 10 hours actually pretty good.
So I had a tweak and added all the mods on Just Blairs site except the level 2 power supply. I then let it run a further 24 hrs. Wow, the sound was smooth, detailed, bass was good, I was impressed. Frankly It sounds better than my gainclones and my clones are actually much better than the std clone so that it quite a hard thing for me to say.
Anyway disaster was just around the corner as my next step was to split the power supply and run the thing as a dual mono of two batteries. In my haste and evening blindness I connected one of the earths to the mute earth, instead of the real GDN.
On firing up the little beasty I heard a pop of sorts and then lots of distortion on the channel that was incorrectly hooked up, the other channel is perfect.
Not to worry at $20.00 the learning value was fine and I will buy a few more boards including the 2 by 100w ones for the bass drivers.
One thing was clear this amp has amazing potential and I will be using these instead of the gainclones. I will post some ideas I have in the next post.
Recently I bought the little sure board as I was interested to see how it performed compared to my gainclones. I need several amp modules to complete my open baffle system and was about to embark on building my final gainclone modules but felt that digital modules might be a good option as my whole system is "off the grid" and it is much easier to use a single sided supply rather than the dual rail supplies needed for the gainclones.
When I first fired up the 2024 I was not much impressed, it sounded thin, but I knew it takes a bit of burn in so I let it run, within just half an hour it was sounding better and by 10 hours actually pretty good.
So I had a tweak and added all the mods on Just Blairs site except the level 2 power supply. I then let it run a further 24 hrs. Wow, the sound was smooth, detailed, bass was good, I was impressed. Frankly It sounds better than my gainclones and my clones are actually much better than the std clone so that it quite a hard thing for me to say.
Anyway disaster was just around the corner as my next step was to split the power supply and run the thing as a dual mono of two batteries. In my haste and evening blindness I connected one of the earths to the mute earth, instead of the real GDN.
On firing up the little beasty I heard a pop of sorts and then lots of distortion on the channel that was incorrectly hooked up, the other channel is perfect.
Not to worry at $20.00 the learning value was fine and I will buy a few more boards including the 2 by 100w ones for the bass drivers.
One thing was clear this amp has amazing potential and I will be using these instead of the gainclones. I will post some ideas I have in the next post.
Gainclone concepts and the Sure board
I know lots of folk run the sure boards on switchmode supplies and I have not tried that on the Sure amp but I tried switchmodes, linear and battery supplies on the gainclones, preamps, DACs, phone preamps and other digital devices.
I must point out my speakers are highly efficient at around 98db, they basically lay bare the signal fed to them so resolution, low distortion and low circuit noise are paramount. In every case I found batteries were best at any point in the chain and RAW DC at that, putting a regulator in the circuit always seemed to degrade things.
I suspect that in situations like mine the same will apply to the D amps.
Switchmodes were the second best in my gainclones, at first they sounded a bit more dynamic and zingy in the highs but I realized after a while what I was hearing was a type of distortion that was colouring the sound. Still the switchmodes killed the linear supplies stone dead.
Something subtle I found with Gainclones was that getting the batteries as close as possible to the chips with as thick a wire as I could get into the system made a difference to bass drive and overall dynamics, once again I suspect the Sure boards will respond similarly and in my test board I had only 3 or 4 inches of large gauge cable between the board and the battery.
Bass was very good even without extra caps, though I added a pair of 2000uf caps at the power inputs on each side of the board.
Generally the above approach meant that I didn't need to add too high a capacitance to the power rails on the gainclones which is important as very high levels of C kills the high end resolution on most gainclones, not sure how it might effect D amps though.
Gainclones benefit for very compact point to point construction with perfect star earthing, this sweetens the amp and increases resolution in my experience, especially with the musical LM1875 chips.
I suspect the D amps might also benefit from similar approaches, making the board smaller is out but Star earthing is possible. My idea is to take the power supplies direct to the positive rail near the chips, splitting the rail in the centre as per Blairs site. Run a large cap from each of these points to the earth plane beneath the chip plus a small tant de-coupling cap. Attach all earth points to a star point underneath the chip, including the speaker returns (after circuitry of course), inputs GNDs and the power. Then cut all redundant traces.
Run air core inductors housed in copper plumbing tubes with caps and shield both sides of the boards with copper clad PCB boards earthed to the star point.
My tests have shown that a cap of about 3000uf across the battery supplies also helps as it kills residual noise from the battery, the value seems to be pretty specific to the battery however.
The size of the battery also seems to have an effect on gainclones and DACs, larger batteries just work better, with big old car batteries really rocking, but having them in the lounge room is not a great idea. On my 2024 test board I was using a 18 ah battery so that may have had a part in the great sound I was getting. I suspect a few folk claiming average results from battery supplies may have been using weedy batteries with long connection cables and thin wires.
All the above might sound like overkill but I suspect it might just elevate the board to another level for not much outplay as generally the higher the resolution of the device the better they respond to small changes and there is no doubt the 2024 is a high res device.
I know lots of folk run the sure boards on switchmode supplies and I have not tried that on the Sure amp but I tried switchmodes, linear and battery supplies on the gainclones, preamps, DACs, phone preamps and other digital devices.
I must point out my speakers are highly efficient at around 98db, they basically lay bare the signal fed to them so resolution, low distortion and low circuit noise are paramount. In every case I found batteries were best at any point in the chain and RAW DC at that, putting a regulator in the circuit always seemed to degrade things.
I suspect that in situations like mine the same will apply to the D amps.
Switchmodes were the second best in my gainclones, at first they sounded a bit more dynamic and zingy in the highs but I realized after a while what I was hearing was a type of distortion that was colouring the sound. Still the switchmodes killed the linear supplies stone dead.
Something subtle I found with Gainclones was that getting the batteries as close as possible to the chips with as thick a wire as I could get into the system made a difference to bass drive and overall dynamics, once again I suspect the Sure boards will respond similarly and in my test board I had only 3 or 4 inches of large gauge cable between the board and the battery.
Bass was very good even without extra caps, though I added a pair of 2000uf caps at the power inputs on each side of the board.
Generally the above approach meant that I didn't need to add too high a capacitance to the power rails on the gainclones which is important as very high levels of C kills the high end resolution on most gainclones, not sure how it might effect D amps though.
Gainclones benefit for very compact point to point construction with perfect star earthing, this sweetens the amp and increases resolution in my experience, especially with the musical LM1875 chips.
I suspect the D amps might also benefit from similar approaches, making the board smaller is out but Star earthing is possible. My idea is to take the power supplies direct to the positive rail near the chips, splitting the rail in the centre as per Blairs site. Run a large cap from each of these points to the earth plane beneath the chip plus a small tant de-coupling cap. Attach all earth points to a star point underneath the chip, including the speaker returns (after circuitry of course), inputs GNDs and the power. Then cut all redundant traces.
Run air core inductors housed in copper plumbing tubes with caps and shield both sides of the boards with copper clad PCB boards earthed to the star point.
My tests have shown that a cap of about 3000uf across the battery supplies also helps as it kills residual noise from the battery, the value seems to be pretty specific to the battery however.
The size of the battery also seems to have an effect on gainclones and DACs, larger batteries just work better, with big old car batteries really rocking, but having them in the lounge room is not a great idea. On my 2024 test board I was using a 18 ah battery so that may have had a part in the great sound I was getting. I suspect a few folk claiming average results from battery supplies may have been using weedy batteries with long connection cables and thin wires.
All the above might sound like overkill but I suspect it might just elevate the board to another level for not much outplay as generally the higher the resolution of the device the better they respond to small changes and there is no doubt the 2024 is a high res device.
Sure, this is not Sure🙂 this is much , much better - Arjen's new board!
However little boring for DIYers - build so perfect, no space for mods🙂
Hey Zigus that board certainly looks the business, I checked them out on Arjens site and have decided to try one, I will run mine of a large battery.
If it can sound better than the modded sure did then it must be very impressive indeed, how long does it take to run in properly?
If it can sound better than the modded sure did then it must be very impressive indeed, how long does it take to run in properly?
Ta 2020 Boards
I also have one of Arjens MkIII boards. Very nice board. I found it slightly sharp to my personal taste & changed the input caps to Mundorf M Caps.
I also changed C13 & R01 on left channel & equivalents on right channel to .22 uf & 10 ohm respecively. Those mods took it a wee bit too far the other way so I may restore the output mods. I am using bright speakers, AR M1's., & a linear 20 amp PS. It has the best soundstage of any of the Tripath amps I have tried. SI, Sure, Charlize & Arjens 2 earlier boards.
Paul
I also have one of Arjens MkIII boards. Very nice board. I found it slightly sharp to my personal taste & changed the input caps to Mundorf M Caps.
I also changed C13 & R01 on left channel & equivalents on right channel to .22 uf & 10 ohm respecively. Those mods took it a wee bit too far the other way so I may restore the output mods. I am using bright speakers, AR M1's., & a linear 20 amp PS. It has the best soundstage of any of the Tripath amps I have tried. SI, Sure, Charlize & Arjens 2 earlier boards.
Paul
- Status
- Not open for further replies.