[ClefChef]

Here's one I have just finished. It is based on modified Audiobah board, has two inputs, everything is controlled by remote - on/off, volume, input selection. Another identical one is in the pipeline so this may become available for sale.

Audiobah is the best sounding board I have used. Modifications are limited to replacing power supply caps with 2200uf and bypassing them with 3.3uf wimas underneath. There is also a large power supply cap on the power input (next to power relay).

Everything works and sounds just beautiful. I will put dry transfer lettering and logos later.

[dboy]

Lovely neat layout Clefchef - I wish my work was so tidy. Did you really get a couple of 2200uF caps onto that Audiobah board, or did you mean 220uF? I am curious what type of caps you used?

I noticed that you stuck to using the onboard connectors. Without ever actually doing listening tests, I have always assumed it is better to replace these with soldered joints for the final installation in a box. I guess the performance advantage is minimal, I just always went on the idea that all connectors are flawed to some extent and that soldered joints are less so (eg. you can hear the differences between RCA sockets, binding posts etc.) If anyone can contradict me here, that might make my life marginally simpler in the future. 

[Poultrygeist]

Outstanding! ClefChef.

Hope I get to hear one before you sell them all.

[ClefChef]

Lovely neat layout Clefchef - I wish my work was so tidy. Did you really get a couple of 2200uF caps onto that Audiobah board, or did you mean 220uF? I am curious what type of caps you used?

I noticed that you stuck to using the onboard connectors. Without ever actually doing listening tests, I have always assumed it is better to replace these with soldered joints for the final installation in a box. I guess the performance advantage is minimal, I just always went on the idea that all connectors are flawed to some extent and that soldered joints are less so (eg. you can hear the differences between RCA sockets, binding posts etc.) If anyone can contradict me here, that might make my life marginally simpler in the future. 

I did indeed put in a couple of Nichicon 2200uf caps. I listened for a bit and decided to replace them with 370uf Panasonic Hybrids that are on the way from Mouser. The quality of capacitors vary, especially when the source is eBay, so listening test is the ultimate test for them.

Onboard connectors are a must in a modular construction - it allows for quick servicing without soldering and de-soldering connections. Performance gains are somewhat negligible  Just look inside your speaker - do you see wires soldered to the drivers and to binding posts?

[shadowlight]

ClefClef, any information on the remote control and the chasis that you are using with the mod?  Source that you purchased it from would be helpful.

[ClefChef]

ClefClef, any information on the remote control and the chasis that you are using with the mod?  Source that you purchased it from would be helpful.

eBay is your friend 

[dboy]

Performance gains are somewhat negligible  Just look inside your speaker - do you see wires soldered to the drivers and to binding posts?

I certainly do... but that's 'cos I made 'em! 

[dboy]

the other much cheaper greenboard (ebayprices) has a much shorter distance from chip to decoupling ceramics.

Based on observation above by Markvdv I purchased the board (I believe it is known as 'YBDZ' perhaps someone can confirm?) Tested it yesterday and although it had a truly horrendous turn on pop/fart noise, the sound quality exceeded Audiobah quite comfortably. It was noticeably more detailed and refined - despite costing less than 15 USD.

I decided to use this board to replace the "look no caps" Audiobah board that I have been integrating with a Bybee Music Rail. I removed the tiny SMD input caps (it will mate exclusively with my Buffalo DAC which has isolating transformers as the output stage). I soldered some silver wire to the freed-up SMD pads nearest the chip pins (4,5,10 & 11) to take the input signal.

Because of the stability requirement advised by Bybee (total capacitance across power rails >33uF & <470uF) I removed the 3x470uF electrolytics onboard and replaced with 2x47uF 16v Black Gate FK left over from an old project. I do not need this capacitance for bass, just staying within the stability requirement. I fed power+ from the MR to an area of exposed metal between the chip and the electrolytic and power- to the unused signal gnd connection. This gave a nice short path from MR to chip.

Connected up to the Buffalo and it does sound noticeably better than the already excellent "look no caps" amp. The sound is softer, more intimate, more enveloping. Soundstage is enormous. Also the turn-on pop/fart is gone. I may or may not get around to changing the inductors on this board (and I may still fiddle with other amps for entertainment) but this is basically my system for the next five years or so now. It is by far the best I have ever heard a stereo sound, and I'm gonna relax a bit now and enjoy the music. Cheers to anyone - especially Markvdv - who offered advice. Sorry for all the long posts! 

[ClefChef]

Based on observation above by Markvdv I purchased the board (I believe it is known as 'YBDZ' perhaps someone can confirm?) Tested it yesterday and although it had a truly horrendous turn on pop/fart noise, the sound quality exceeded Audiobah quite comfortably. It was noticeably more detailed and refined - despite costing less than 15 USD.

The "truly horrendous turn on pop/fart noise" is not synonymous with quality - I wouldn't use this board unless fitted with speaker protection modules. Somehow I think that you may have killed your Audiobah board completely and anything else may sound better now 

[dboy]

The "truly horrendous turn on pop/fart noise" is not synonymous with quality - I wouldn't use this board unless fitted with speaker protection modules. Somehow I think that you may have killed your Audiobah board completely and anything else may sound better now 

As I said, the pop is gone. I believe it was a product of a mismatch of input caps (now removed). The Audiobah board is not killed, and if you check the post below you will see that I spent the weekend setting up a different Audiobah board for a friend in a more conventional amp - i.e. just replacing the ecaps for Elna Cerafine and adding a big Mlytic cap near the power input. A setup that is similar and probably sounds much like the one you posted above. Believe me, what I described today sounds a whole world better!!!

P.S...

[Markvdv]

8)dboy. Inputcaps you removed I think were 8uF, that causes pop. But board has connector for SD and mute, a simple switch between SD&mute mutes the amp and between SD and GND puts the amp in standby, both could be used to avoid pop. 8uF value is I think to have lossless bass respons even when gain is set to highest 36dB, it is 26dB standard on Wiener (30C=20K). Transformers are better

Probably very little room to improve further, Wurth inductors might be little better for high frequency detail, polyprop filtercapacitors may also sound a little better. Values chosen for speakers used. Gainsetting 20dB could have little less distortion if inputlevel is high I feel but I haven't compared that on transformers on input. Inputimpedance changes too, there might be a best match setting for the transformers you used.

[dboy]

Hey - there you go giving me more to think about when I want to just relax and spin some tunes! 

You make good points of course. I think especially the matter of input impedance may leave room for a worthwhile tweak to match better with the transformers (Sowter 3575). I have plenty of spare volume. For now though I want to let the system settle down and to get really used to how it sounds so it is easier to hear the effects of any fine tuning. Thanks again for all of your input - I'll report back when the time comes to try your suggestions!

[ClefChef]

After 24 hour burn-in period the amp has opened up quite nicely - it sounds excellent. After all those Nichicon FM 2200uf caps with Wima bypasses underneath work perfect.

[downunder55]

eBay is your friend 

Have spent plenty of time on ebay already looking for a remote solution, before I found your post of July 14th with the great picture.

Had found this for vol only:  http://www.ebay.com.au/itm/6CH-Remote-Motor-ALPS-Volume-Potentiometer-Control-Adjust-50K-6-W-Remote-/201077764009?pt=LH_DefaultDomain_0&hash=item2ed12b2fa9&_uhb=1

There are plenty with multiple inputs (4-6), but yours seems to only have two which is ideal, can you please help us out with a source ?

[ClefChef]

Have spent plenty of time on ebay already looking for a remote solution, before I found your post of July 14th with the great picture.

Had found this for vol only:  http://www.ebay.com.au/itm/6CH-Remote-Motor-ALPS-Volume-Potentiometer-Control-Adjust-50K-6-W-Remote-/201077764009?pt=LH_DefaultDomain_0&hash=item2ed12b2fa9&_uhb=1

There are plenty with multiple inputs (4-6), but yours seems to only have two which is ideal, can you please help us out with a source ?

There are plenty of 2 channel relay boards on eBay, you would have to modify them a bit. The remote you posted has 4 input selector + on/off channel. To use 2 input channels you have to bridge 1+3 and 2+4 on the remote. The remote puts out +5V signal on the input select channels, so you could either use 5V relay boards and drive relays directly (remote provides enough current for that) or use 12V relay boards and drive them via on-board trigger circuitry.
On/off channel is a "shunt" channel (board controls connection to ground) so your voltage there depends on the positive voltage provided (e.g. drive 12V on/off relay with +12V or 24V relay with +24V, etc.) You would use the on/off channel relay to control the power supply to your TPA3116 board.

On my amp you can see a LT7812 (attached to the bottom next to the relay) to provide 12V for the remote board and the on/off relay (next to the 7812). 7812 also protects the remote board from your main power source which is going to be somewhere between 18-24V. Remote board has its own LT7805 so you can feed it 9-15V.

Here's the relay board similar to what I used: http://www.ebay.com/itm/400272908552?_trksid=p2059210.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

You need to parallel the outputs, remove all the resistors and transistors and connect the remote board switching signal wires directly to relay coil legs.


I blew up the picture for greater detail on the remote board. The channel control pins are marked Q4-Q8 on the board with Q9 being ground pin. In order to have 2 instead of 4 inputs provided by the board you have to bridge Q4 with Q6 and Q5 with Q7 (blue lines), and use the +5V and Ground signal to control the input selection relays.

[downunder55]

There are plenty of 2 channel relay boards on eBay, you would have to modify them a bit. The remote you posted has 4 input selector + on/off channel. To use 2 input channels you have to bridge 1+3 and 2+4 on the remote. The remote puts out +5V signal on the input select channels, so you could either use 5V relay boards and drive relays directly (remote provides enough current for that) or use 12V relay boards and drive them via on-board trigger circuitry.
On/off channel is a "shunt" channel (board controls connection to ground) so your voltage there depends on the positive voltage provided (e.g. drive 12V on/off relay with +12V or 24V relay with +24V, etc.) You would use the on/off channel relay to control the power supply to your TPA3116 board.

On my amp you can see a LT7812 (attached to the bottom next to the relay) to provide 12V for the remote board and the on/off relay (next to the 7812). 7812 also protects the remote board from your main power source which is going to be somewhere between 18-24V. Remote board has its own LT7805 so you can feed it 9-15V.

Here's the relay board similar to what I used: http://www.ebay.com/itm/400272908552?_trksid=p2059210.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

You need to parallel the outputs, remove all the resistors and transistors and connect the remote board switching signal wires directly to relay coil legs.


I blew up the picture for greater detail on the remote board. The channel control pins are marked Q4-Q8 on the board with Q9 being ground pin. In order to have 2 instead of 4 inputs provided by the board you have to bridge Q4 with Q6 and Q5 with Q7 (blue lines), and use the +5V and Ground signal to control the input selection relays.

Wow, thankyou so much for the great detail, was looking specifically for a remote kit with only the 2 inputs, I see your approach, ......now I have a lot to read and digest in your post .... thanks muchly 

[Poultrygeist]

ClefChef's the man!

[ClefChef]

I did some more experimentation on one of my Audiobah amplifiers - I replaced SMD signal coupling 1uf caps with 3.3uf wima. There are two input caps and two AC to ground caps, all four have to be identical. On the board these caps are marked C15, C16, C18, and C7. After replacing signal coupling caps with bigger ones you also need to replace a reservoir cap in the startup mute circuit (C5). This capacitor sets the mute interval that blocks turn-on pop while the input capacitors are charging, original value seems to be more than 4.7uf (the limit on my Fluke tester  ). Now that we have larger input caps the mute interval becomes too short and turn-on pop will go boom through your speakers. I used 47uf lytic and that seems to have done the trick - amplifier un-mutes couple of seconds after applying power, no pops - no noises.

Wimas sound a lot better, no doubt.

The procedure requires fine point soldering iron and steady hands 

Audiobah board seems to be easily convertible to differential inputs (balanced), which would be worthwhile in my case. I was going to build two small TDA3116 mono-block amplifiers with ferrite bead output filters and 1' speaker cables that would attach directly to speaker binding posts (cling-on amplifiers). Balanced connection in such scenario is a must in my opinion.

[Markvdv]

Look at output and what happens there on "audiobah".
Voltage coefficient of smd ceramics makes using them in amplifier output(filter) tricky, next to that there is distortion to consider when chosing the correct ceramic smd there. I don't feel you can reasonably expect smd ceramics in output on a $8 ampboard to be close to optimal. Using filmcap there like many other manufacturers do is an easier and wiser choice IMO.

[dboy]

Gainsetting 20dB could have little less distortion if inputlevel is high I feel but I haven't compared that on transformers on input. Inputimpedance changes too, there might be a best match setting for the transformers you used.

Although I wasn't going to tweak for a bit, I couldn't resist trying this. Not that there was any problem with distortion or hiss (speakers silent with ear up close). But it makes sense that a higher input impedance would work better with the transformers (10k:10k nominally).

The eBay listing I bought the board from said gain was 36db, but the resistor (R2) that I removed from the board was 100k, which concurs with your (markvdv) assertion of 26db. I left the other resistor unchanged on assumption that the value did not have to match with datasheet value since it is a voltage divider and R2 is left open for 20db. I think I have read that this is OK?

Anyway, gain does seem reduced. The bass is stronger and the sound generally is weightier, although dynamics may be slightly less. Still, there is no shortage of dynamics. Overall, this is definitely an improvement. So if anyone else is tempted to try input transformers, my experience suggests to go for the lowest gain.