http://www.hypex.nl/docs/papers/The G Word.pdf
this is not a beginner project as Surface mount, but does give as good performance as you can probably get and is balanced. Also a good read on understanding what balanced actually means in a way that you can grasp.
this is not a beginner project as Surface mount, but does give as good performance as you can probably get and is balanced. Also a good read on understanding what balanced actually means in a way that you can grasp.
I'm quite familiar with red book CDs. One of my favorite test albums, Dire Straits "Brothers in Arms" is recorded a good 10 dB below clipping and requires some cranking of the volume control. If I really crank it, I may reach -10 dB on the preamp with that album.
If your source is a phonograph or cassette tape player, you'll need some gain in the system. To preserve SNR, put this gain as early in the signal chain as possible. I.e. build a preamp with gain.
~Tom
I've been thinking about preamps for a while, so I'd say there is a plan but no real action just yet...
Right now, I'm pushing on the Parallel-86. Once that's done I have a couple of small projects that I'd like to get done. Then I may have some bandwidth to think more about preamps.
One of my smaller projects could actually be used as a simple preamp by adding a volume pot.
Stay tuned...
~Tom
There's a pretty easy way to find out. See the blurb about determining the thermal resistance of a heat sink near the top of my Taming the LM3886 - Thermal Design page.
One of those heat sinks per channel is probably OK. That's my gut feel.
~Tom
This album, http://www.discogs.com/Kiri-Te-Kana...a-Verdi-Puccini-Verdi-Puccini/release/5043478, is one of the worst "offenders". Plus, it has faulty mastering at the beginning of the first track - you can hear digital gurgling being quite prominent in the noise floor.
Thanks for the CPR! And the advice. I'll probably get back to you with questions. I've read your other posts and am curious to know how your M86 project is panning out.
Great emoticon! Took me a bit to figure it out.Complete Newbie!
...
Likewise, thanks for the advice and the detailed pointers. I'll check out the information and get back to you.
This sounds appealing, and seems to match the goal of minimizing unnecessary gain. Or am I exposing my ignorance? Either way, I'll try to figure out a circuit and then ask for suggestions.
Since I have bought the boards, tapping into the 15V power supply is a no brainer, no?
You've earned it, I think!(Pardon the MarketingSpeek™. It's that kind of morning...)
I have a lot of reading and learning to do based on the three quoted posts. In the meantime, I'd like to offer up my opinion as a relatively new board reader. There are two threads on the Modulus86 that I've been following; this one and the one over on the vendor forums. Both of these have approximately the same content spectrum: a mix of technical discussions, arguments, philosophy and actual M86 related build details and experiences.
I just spent a few hours going over every post so as not to miss any information, and most of it went flying way over my head.
Wouldn't it be great if there were three threads:
Modulus86 Announcements - for Tom to make announcements, updates, etc.
Modulus86 Technical Discussion - for people to analyze, criticize, etc.
Modulus86 Build Experiences - for people to relate their real-life build experiences
And wouldn't it be great if people would actually stick to the topic of the boards. But ah, this is a forum and the above is a pipe-dream...
Should be pretty straight forward. You can pick off the ±15 V from D10 and D12. You should be able to pick off a supply ground from C24 (for Modulus-86 Rev. 1.0) or C27 (Modulus-86 Rev. 2.0) as well.
I'd probably use the THAT1646 for the preamp. Single ended in, differential out. Put the volume pot on the input side. Just follow the recommendations given in the data sheet. If you need help, I suggest starting a thread in the Analog Line Level forum. I'll be happy to contribute there as time allows. The occasional question via email is OK as well.
Yeah... The two have basically merged into the Modulus-86 thread in the Vendor's Bazaar forum. I will update Post #1 in the thread whenever there's something major to report. Also, keep an eye on my website - Neurochrome Audio: Precision high performance audio circuits for the DIY market.
That's basically this thread.
Pink ponies, unicorns, and rainbows for everyone....
~Tom
Hello,
Yesterday I received the printed circuit boards (in France).
Today, I have almost finished populating the components on the boards.
Problem: Because of an error (or a bug) from myself or from the BOM, the connectors for the PSU are missing. I have to order them and I should not have them before Tuesday ...
The principle of the double bottom to conceal the wiring takes shape.
Yesterday I received the printed circuit boards (in France).
Today, I have almost finished populating the components on the boards.
Problem: Because of an error (or a bug) from myself or from the BOM, the connectors for the PSU are missing. I have to order them and I should not have them before Tuesday ...
The principle of the double bottom to conceal the wiring takes shape.
Attachments
Yay!
That's pretty weird. They are on the Mouser project. I just checked... There must have been a bug somewhere.
Also conceals the head of the power transformer mounting bolt. I like that. Your build is shaping up very nicely. Good work.
~Tom
Congratulations! Looking good. Do let us know how you like the sound. I'm building one too, but I'm much slower than you...
Incidentally, I also had a problem with the number of PSU connectors (I was 2 short). I was also missing the C9 capacitor. (I was ordering from the US, though.)
That's a neat case. Could you provide a link?
I think I know what's going on. C9 (MOD86 R1.0), Mouser P/N 80-R82EC3100AA70J is used both in the power supply and the amplifier. Same for the 3-terminal connectors.
I just tried placing 2 x MOD86 R1.0 and 1 x PWR86 in my cart at Mouser. It does indeed miss the connectors for the power supply and C9 for the amp. This is a bug in Mouser's system. Grr...
I'll address this with Mouser. Thanks for bringing this issue to my attention.
When you notice discrepancies like this, please shoot me an email ASAP so I can minimize the damage.
~Tom
I just tried placing 2 x MOD86 R1.0 and 1 x PWR86 in my cart at Mouser. It does indeed miss the connectors for the power supply and C9 for the amp. This is a bug in Mouser's system. Grr...
I'll address this with Mouser. Thanks for bringing this issue to my attention.
When you notice discrepancies like this, please shoot me an email ASAP so I can minimize the damage.
~Tom
Tom, thanks for your prompt attention.
I actually thought it was something I did. It was only when Dom reported exactly the same problem that I started thinking it may be systematic.
Finally assembled the boards and tried them out. It lives! This is my first amp build and I didn't have to redo anything! Yet.
For those thinking about trying this out, here is my experience so far (in the US). The boards arrived in a few days, the transformer (AS-2222, Antek) and the Mouser package in a week. I was bitten by the bug in the Mouser system (see Post #334) which causes two connectors and a capacitor (per channel) to be missing; watch out for this if you order the Power86 and the Modulus86 BOM together from Mouser.
The documentation from Tom is outstanding. Even a clueless newbie like me was able to put together the system in one go. The testing procedure suggested by Tom is simple and since it went without a hitch for me, easy to do.
I put together a Power86 and two channels of Modulus86 in a makeshift configuration and tried it out. There are obviously several weaknesses in my configuration that even I can see. I'll list them here:
1) There is no chassis and nothing is bolted down.
2) I've just wired RCA connectors to the Modulus86 J2 connectors, shorting -ve and ground on J2. Also, I used ordinary wires, no shield, between the RCA connectors and the Power86 board.
3) I used some temporary lightweight heatsinks (from an old power supply). I did use thermal paste.
4) I used non-remarkable speakers so far (for fear of blowing something out
5) I used a crappy source - my cell phone earphone output via an adapter cable.
Despite all the above, it played music, and pretty well.
Now the questions:
1) The transformer has an annoying buzz. I had to put my ear close to it to localize the source, but it's audible in the room. I'm hoping this will go away when I bolt it down in a chassis. This is audible without connecting the Mod86 boards; it's not through the speakers, it's directly from the transformer.
2) The doc says that the output DC offset should be below 20uV. I measured 1.1mV on one channel and 0.9mV on the other. (I recall having seen something about this earlier on the forum, but couldn't find it again.) The measurements were taken with the inputs open, not shorted.
3) The sound volume was relatively low. This may be because of my cell phone being the source. But even with the volume on the cell phone turned up all the way, the sound was , well, a quiet listen.
4) I had the feeling that the vocals were somewhat recessed and the instruments somewhat forward, especially the bass. Again, the source, the speakers and the room are far from ideal, but perhaps someone has some explanation independent of these.
And what would a report be without pictures of the war zone. Gentlemen, feel free to comment, criticize, but do avoid direct abuse...
For those thinking about trying this out, here is my experience so far (in the US). The boards arrived in a few days, the transformer (AS-2222, Antek) and the Mouser package in a week. I was bitten by the bug in the Mouser system (see Post #334) which causes two connectors and a capacitor (per channel) to be missing; watch out for this if you order the Power86 and the Modulus86 BOM together from Mouser.
The documentation from Tom is outstanding. Even a clueless newbie like me was able to put together the system in one go. The testing procedure suggested by Tom is simple and since it went without a hitch for me, easy to do.
I put together a Power86 and two channels of Modulus86 in a makeshift configuration and tried it out. There are obviously several weaknesses in my configuration that even I can see. I'll list them here:
1) There is no chassis and nothing is bolted down.
2) I've just wired RCA connectors to the Modulus86 J2 connectors, shorting -ve and ground on J2. Also, I used ordinary wires, no shield, between the RCA connectors and the Power86 board.
3) I used some temporary lightweight heatsinks (from an old power supply). I did use thermal paste.
4) I used non-remarkable speakers so far (for fear of blowing something out
5) I used a crappy source - my cell phone earphone output via an adapter cable.
Despite all the above, it played music, and pretty well.
Now the questions:
1) The transformer has an annoying buzz. I had to put my ear close to it to localize the source, but it's audible in the room. I'm hoping this will go away when I bolt it down in a chassis. This is audible without connecting the Mod86 boards; it's not through the speakers, it's directly from the transformer.
2) The doc says that the output DC offset should be below 20uV. I measured 1.1mV on one channel and 0.9mV on the other. (I recall having seen something about this earlier on the forum, but couldn't find it again.) The measurements were taken with the inputs open, not shorted.
3) The sound volume was relatively low. This may be because of my cell phone being the source. But even with the volume on the cell phone turned up all the way, the sound was , well, a quiet listen.
4) I had the feeling that the vocals were somewhat recessed and the instruments somewhat forward, especially the bass. Again, the source, the speakers and the room are far from ideal, but perhaps someone has some explanation independent of these.
And what would a report be without pictures of the war zone. Gentlemen, feel free to comment, criticize, but do avoid direct abuse...
Attachments
Last edited:
It's weird with the transformer hum. I think I've only heard that with E-I core transformers. Toroids are usually dead quiet. I'd toss Antek an email. I've used Antek transformers both in solid state and vacuum tube builds and never had issues with mechanical hum.
The 20 µV mentioned in the doc is a bit optimistic. I generally see around 150-200 µV. The 1 mV you're measuring is a bit much. You do need to measure with the inputs shorted, though.
~Tom
The 20 µV mentioned in the doc is a bit optimistic. I generally see around 150-200 µV. The 1 mV you're measuring is a bit much. You do need to measure with the inputs shorted, though.
~Tom
Last edited:
The transformer noise is likely to become more audible when inside the resonant structure of a metal Chassis.
There are two things you should try.
1.) Try adding a DC blocking capacitor bank in the Primary feed to the transformer.
2.) Investigate the change in primary current with mains voltage change.
If the first works,then you don't need the second.
Near saturation of a 220Vac (or 110Vac) transformer running on 230/240Vac (or 115/120Vac) usually results in excess vibration of the transformer components.
3.) Try dipping the warmed transformer into a bath of transformer varnish. Then leave to dry thoroughly. This may need to be repeated one or two more times to get the varnish into the smallest of the internal air gaps.
This helps stop the core laminations and the winding wire vibrating in sympathy with the mains current pulses.