Looks like I will be preordering... With 3 old amps sitting around, it would be interesting to replace the guts of a problematic one first. Given the limited time I have, I would love to go with the assembled boards.206 W into 8 Ω @ 0.0006 % THD+N verified. 350 W (4 Ω) when my lab supply hits its current limit. I expect the actual limit to be closer to 400 W.
6 x LM3886 + LME49724 + OPA1642 in a Bridge/Parallel Differential Composite Amp architecture.
Availability: Preorders by end of January 2018. Full production towards mid/late February.
Just saying...
Tom
that would require 38Vpk, which is just about impossible from a group of paralleled lm3886 supplied from a regulated +-41.99Vdc.
To get 90W into 8ohms the Rowlands would almost certainly have used BPA300 topology.
Last edited:
If you back up to the writing (page 2) about this amp you'll discover its not based on the LM3886 at all. Here's part of the description :
The output section features four individually selected bipolar transistors per phase per channel, totaling 16 transistors overall.
I believe the Jeff Rowlands using 3886's were the Concentra integrated (first 2 pics) and the model 10 (second 2 pics) which had a separate regulated psu... there may have been other but those are the ones I remember. Looking at the claimed specs - Concentra 100W/8ohm, 150W/4ohm and Mod 10 150W/8ohm, 275W/4ohm...
John
John
Attachments
JRDG Company Special – Model 10, 12 and 112 | HFA - The Independent Source for Audio Equipment Reviews has some more information on the amps.
Would make a nice chassis for a mod amp.
Would make a nice chassis for a mod amp.
Not impossible at all. With six LM3886es in parallel, each see a 48 Ω load, effectively. Thus, the dropout across the LM3886 is pretty small.
I do agree that this would require a regulated supply that provides 42.000 V ABS MAX with very tight tolerances, but it is within the realm of possibility.
I would openly question the sanity of anyone attempting such a feat, though. There really are better ways to get power from an LM3886. There I think we do agree.
From the chatter here, it appears I'm barking up the wrong tree. I made some assumptions about Jeff's amps without reading all the detail. My bad!
I have no desire to reverse-engineer his amps.The MOD686 will compete with the best amps in the world and giving a world-class amp design away for cheap does not seem like a good business model.
I do have 7-8 of the prototype boards (shown in the build pic I sent out on Christmas eve) that I will consider offering up to the world. At this point, it looks like the final build will be the exact same board. The only difference will be that the final revision will be gold plated (ENIG) for better automated solder ability. I'll announce the opening of pre-orders and any sales through my newsletter first. You can subscribe to it here: Neurochrome Newsletter.
I'm seriously thinking about taking a page from Schiit's playbook, though. They don't do sales. Why? Because sales make the product more expensive for everyone and results in disgruntled customers. You can read their notes on the topic here: Schiit Sales. Sales are only good business if the added sales volume makes up for the reduction in revenue from the lower sales price during the sale. DIY is an extremely low-volume business. It doesn't lend itself well to sales.
Tom
Last edited:
Not to deviate too much from topic, but where do you find those bus strips like Rowland used? In my previous life we used ones without the insulated coating. I've not been able to locate either at Digi-key, Mouser, etc. I may have to resort to thumbing through an old catalog looking for the correct picture as searches of the web site have been fruitless.
Tom, I agree with you 100%.
A well designed product sells itself. People often lose sight of the fact that development takes TIME. And money, and resources, and nerves. You also have to make enough profit that the whole venture is, in the end, worth the effort.
I have one of your MOD86 amps as a daily-driver and I am more than satisfied with it. In fact, my tube amps have become dusty because it has been in the rotation so long. A fair price for an outstanding product. What more could one ask for?
I for one will be eagerly awaiting the new release and will certainly be excited to start building!!
We are in end effect not just paying for parts and a PCB, but for your expertise, time, talent, and your desire to provide the DIY community with a better than average product.
I'm game.
Thanks again, -Mac
A well designed product sells itself. People often lose sight of the fact that development takes TIME. And money, and resources, and nerves. You also have to make enough profit that the whole venture is, in the end, worth the effort.
I have one of your MOD86 amps as a daily-driver and I am more than satisfied with it. In fact, my tube amps have become dusty because it has been in the rotation so long. A fair price for an outstanding product. What more could one ask for?
I for one will be eagerly awaiting the new release and will certainly be excited to start building!!
We are in end effect not just paying for parts and a PCB, but for your expertise, time, talent, and your desire to provide the DIY community with a better than average product.
I'm game.
Thanks again, -Mac
These guys make a product like them:
E-FAB Electronics Bus Bars Copper Busbars & PCB Stiffeners | Photo Etching, Photo Chemical Machining & Micro Milling
Rogers Corporation also make something like that (ROLINX).
Nice!206 W into 8 Ω @ 0.0006 % THD+N verified. 350 W (4 Ω) when my lab supply hits its current limit. I expect the actual limit to be closer to 400 W.
6 x LM3886 + LME49724 + OPA1642 in a Bridge/Parallel Differential Composite Amp architecture.
Availability: Preorders by end of January 2018. Full production towards mid/late February.
Just saying...
Tom
Although I do remember saying that for more power than the Parallel-86 you'd be going discrete...
Not the exact same thing, but Protocase makes bus bars: Copper Bus Bars, built in 2-3 days
They can even powder coat them for you.
Tom
Good to hear.
Yep. Running a business takes skill, time, money, and balls. It's easy to be the Monday morning quarterback. It's rather different when you're the one forking over $10k of your own money to get a short production run started, not knowing when you will recover your investment - assuming the product sells.
I'm not complaining. I did purposely quit my engineering job to go do Neurochrome (and more schooling) full time. That was my decision and I am much happier now ... but I did take a 60-70 % salary cut by doing so.
Thank you. I really appreciate it. It's feedback like that, which keeps me going and makes me want to take the business risk.
The irony is not lost on me...
I used to hammer the BPA designs pretty hard. If you follow any of the BPA design/build threads, it quickly becomes apparent that the BPA topology comes with its own slew of challenges. Turning that further into a B/P composite opens up a giant can of worms. I keep learning and keep finding new ways to solve or work around the various issues, so I have gradually warmed up to the idea of a B/P composite amp.Writing up the Output Power section of my Taming the LM3886 pages was a nice exercise. I wrote the page on the flight home from my summer holiday last summer. The context change was nice and made me see things a bit differently. It provided the proverbial, "oh, well, DUH!" moment necessary to start playing with the MOD686 in the simulator.
Tom
