The reports of listening experiences have started to trickle in. There are a couple in the Modulus-86 Build Thread starting at Post #62.
~Tom
~Tom
The highest power you can get from the Modulus-86 - or any other LM3886-based amplifier - is about 50 W into 8 Ω on ±35 V rails. Note, however that if you choose to operate the amp with a 4 Ω load, it will not be able to deliver the full output swing as the max. output current is 7 A (worst case). You'll need two channels in parallel for that.
If you want more output power than 50 W (8 Ω), you need to bridge two amp channels.
I am currently working on a Parallel-86 board, so a composite amplifier with two channels of "LM3886" in parallel. I put "LM3886" in quotes as I'll actually use the two channels of an LM4780 to do the job. This will allow you to operate on ±35 V rails into 4 Ω loads (and below). This should provide about 50 W into 8 Ω*and about 100 W into 4 Ω. If you really need the extra power, I suggest hooking two Parallel-86 boards in a bridge.
~Tom
Here's one MOD86 builder's subjective comparison of the Modulus-86 and the Sympatico: Modulus-86 Build Thread, Post #170.
~Tom
~Tom
Tom,
using the pair of amps inside the 4780 can be for bridged or parallel or stereo operation.
If used a parallel or stereo when the signals are almost identical, then the currents into the chips are "in phase" and the losses through the pin attachments will be higher than for mono chip operation.
I've suggested in the past, that the better way to use 4780 is for bridged operation, where the currents are shared between the dual polarity supplies.
Then parallel as many of these bridged 4780 to achieve the current capability as required.
using the pair of amps inside the 4780 can be for bridged or parallel or stereo operation.
If used a parallel or stereo when the signals are almost identical, then the currents into the chips are "in phase" and the losses through the pin attachments will be higher than for mono chip operation.
I've suggested in the past, that the better way to use 4780 is for bridged operation, where the currents are shared between the dual polarity supplies.
Then parallel as many of these bridged 4780 to achieve the current capability as required.
I believe Tom is working on just such a design. Will be useful if I am short of grunt for my 3Ohm ribbons.
Then on the other hand, when two amps are in parallel, each half of the amp only draws half the current. Besides, the LM4780 has more power pins than two LM3886es combined.
Bond wires are pretty rugged. They can typically handle over 1 A each, so in case of high-current devices, there are normally multiple bond wires in parallel. This lowers the bond wire resistance and to some extent the inductance as well. Figuring out the number and location of bond wires is all part of the IC design process. If there was an issue, I'm sure it would have been addressed.
I would also point out that parallel operation of the two channels in an LM4780 is mentioned as an application on Page 7 of the LM4780 data sheet.
~Tom
Congratulations Tom what a professional job. I am very tempted, will have to start saving some cash soon me thinks.
It would be great to compare your boards with the My-Ref FE's i already have. I have asked but still no idea what their THD is, none of the builders appear to have a scope.
It would be great to compare your boards with the My-Ref FE's i already have. I have asked but still no idea what their THD is, none of the builders appear to have a scope.
Thank you very much. I'm rather proud of it myself, actually.
I agree with your comments about the lack of documentation on many other projects. That's one of my major value-adds, actually. The design documentation for the Modulus-86 is about 30 pages long. The one for the Power-86 is not far behind. I present tons of measurements both here and on my website with plans for more to come.
Most builders do not have access to the gear needed to measure the THD. You will not be able to see it on an oscilloscope. You need state of the art gear, such as the Audio Precision APx525 that I have. You might be able to scrape by with a QuaintAsylum QA400.
~Tom
I agree with your comments about the lack of documentation on many other projects. That's one of my major value-adds, actually. The design documentation for the Modulus-86 is about 30 pages long. The one for the Power-86 is not far behind. I present tons of measurements both here and on my website with plans for more to come.
Most builders do not have access to the gear needed to measure the THD. You will not be able to see it on an oscilloscope. You need state of the art gear, such as the Audio Precision APx525 that I have. You might be able to scrape by with a QuaintAsylum QA400.
~Tom
Thanks Tom that makes perfect sense to me. Your Mod 86 is probably the most professionally documented and thought out diy project i've yet come across. I spent quite a lot on boutique components for my last My-Ref build, not sure i want to do that again.
What you have created has obvious commercial appeal, you are quite right to protect it.
What you have created has obvious commercial appeal, you are quite right to protect it.
Received 2 more Mod86 and another Power86 PCBs today. I'm excited to try them bridged with passive crossover, and later as active crossover amps.
Thank you.
The world class performance of the Modulus-86 is obtained using standard components. Regular metal film resistors and film capacitors is all you need. There's no need for boutique components. Last I checked the component kit at Mouser ran about $45 per channel.
Nice! Thanks again for your repeat business.
~Tom
This is an excellent point. If I were you, I would ask Tom, if he would accept your My-Ref FE amplifier shipped to him for measurements. That's really the only way to know. I would do the same for those who own the Twisted Pair Sympatico design, honestly.
I am planning on shipping my bridged Modulus 86 to Tom (shipping is on me), only because, to my knowledge, he has not measured a bridged version as of yet. And he has not listened to a bridged version as of yet either.
I have measurements of the NCore 400 (which I am also happy to ship to Tom) as well as the UCD 400HG HxR for comparison purposes. Then we can see if our subjective assessments coincide with our objective measurements. It is easy to make sure that the gains of the various amplifiers are closely matched so that output level doesn't sway our bias for one amplifier design over another.
It would be an honest R&D effort. And some fun to boot.
Best,
Anand.
Good point Anand but i live in Scotland and use the My-Ref FE daily, it is an excellent amp imo, the best i've owned. I built mono blocks with a 300va transformer in each. Way more than they need but they never run out of steam in my system, they are quite heavy though.
BCM Bob put a version together to send out to people for evaluation, i wonder if he would send his to Tom for measuring. He is much closer being in the states.
The more comparisons the better, that would give potential builders more to go on.
I like the idea of getting high end sound quality using ordinary components, that and the almost vanishing level of distortion is what got me interested in this project. I know from past experience that low THD does not necessarily mean good SQ but good design can produce both.
BCM Bob put a version together to send out to people for evaluation, i wonder if he would send his to Tom for measuring. He is much closer being in the states.
The more comparisons the better, that would give potential builders more to go on.
I like the idea of getting high end sound quality using ordinary components, that and the almost vanishing level of distortion is what got me interested in this project. I know from past experience that low THD does not necessarily mean good SQ but good design can produce both.
I will be happy to measure any amplifier on my APx525. That is actually a service I have considered offering for a while. What would a fair price be for such a service? I was thinking something like $50 for a basic service (frequency response, THD vs frequency, THD vs power) and maybe something like $80 for a more advanced service that would include IMD and FFT plots. There'd probably be a custom level above that, which would be more like audio consulting + measurements at an hourly rate. Shipping would be additional, of course.
Yeah... That's always a concern. There are also some what-ifs around tube amplifiers and such. Also, what happens if the packaging job is somewhat lackluster and the amp is damaged in shipping... All the fine print in the world won't make the customer feel any better if the USPS decides to deliver the amp by airdrop ... without a parachute.
Sure. I can build a bandpass filter with excellent THD. It will sound ... well ... band limited, though. So stellar THD but not good fidelity. This is why I measure a wide range of parameters and optimize for the best performance I can achieve across all the parameters. The world class performance (both measured and subjective) of the Modulus-86 stem from the stellar PSRR, stellar IMD, flat THD response, and vanishingly low THD. To sound good, a flat frequency response is needed as well. The Modulus-86 delivers that too.
~Tom
Our postal service has in the past damaged two amps and destroyed a set of speakers which i had sold and sent off, they were all well protected and double boxed. They refused to honour the insurance i payed out for because they were not in original packaging. The amps were made by a small firm in England (NVA) who use plain brown cardboard boxes to send them in so they were in original packaging!
It would have taken legal action to get anywhere and thats what they rely on.
They put me right off sending anything fragile of value by post.
It would have taken legal action to get anywhere and thats what they rely on.
They put me right off sending anything fragile of value by post.