Yeah... Keep an eye on that Universal Buffer. Rev. 1.1 of the prototype is in the fab and should be here in a little over a week. I expect that to be the final revision.
The Universal Buffer will feature support for both differential and single-ended inputs (jumper selectable) and provide both differential and single-ended outputs. You can configure the board to have gain if you desire. You can also wire the board to be inverting should you so desire.
It will only be available as an assembled module. Sorry. I can get better performance at a lower price by delivering assembled modules. If my current guess on the price holds, it's only about $20 more than the THAT Driver and THAT Receiver, which it replaces.
Also note that the performance is significantly better than the THAT Driver and THAT Receiver. 137 dB dynamic range! Nearly 140 dB channel separation. Less than 2 uV noise (~20 dB better than THAT TX/THAT RX).
I'll start a separate thread for it in the vendor forum once I have data for the final prototype and know the final pricing. I should know these bits of data before Christmas (this year! 🙂)
Tom
So, Tom, regarding my case describe above, do you suggest, that unballancing on transformer level, adjusting volume with pot and then ballancing again with the board you described is they way to go?
mmm, is SE-DIFF gain also optional? Or does that combination always have gain?
Awesome!
Please add a link here so we're notified, thanks
I'm not sure what you mean by "unbalancing on transformer level".
If you want a preamp with differential input and differential output, I would use one UniBuff to convert from differential to single-ended. Then use a single-ended volume pot. Then another UniBuff to convert to differential.
You can use a balanced/differential volume pot if you wish. Then all you need is a UniBuff on the output (if you want low output impedance). Unfortunately, there are very few balanced/differential volume controls out there, though.
The default (and lowest) gain is 1.0x (0 dB) for all combinations of DIFF/SE input to DIFF and SE output.
If you want higher gain, you add two resistors. They can be through-hole types if you wish.
Will do.
Tom
Happy Thanksgiving Everyone. Not sure if it’s 3 or 4 years now since I’ve completed my build. It plays music or television probably 10-16 hours a day at least, and still sometimes surprises me with how pleasant and detailed it sounds.
I have DSP (volume control) > DAC (SE output) > Tom’s THAT LineDriver (converts SE to Differential) > Modulus86.
Neat to hear that Tom now has something better than THAT Line Driver, but honestly I’m so satisfied that I do not feel an upgrade is necessary.
In my case DSP & DAC are probably the weak point in the system but the m still very pleased. I’ll just continue my current hobby, which has become the search for the best source media (best quality recordings). Very easy to tell the difference between a good v poor recording on my system.
Happy Thanksgiving.
I have DSP (volume control) > DAC (SE output) > Tom’s THAT LineDriver (converts SE to Differential) > Modulus86.
Neat to hear that Tom now has something better than THAT Line Driver, but honestly I’m so satisfied that I do not feel an upgrade is necessary.
In my case DSP & DAC are probably the weak point in the system but the m still very pleased. I’ll just continue my current hobby, which has become the search for the best source media (best quality recordings). Very easy to tell the difference between a good v poor recording on my system.
Happy Thanksgiving.
Last edited:
Tom, I have a question for you. Would you be able to use multiple Modulus-86's in some kind of parallel or bridge configuration to drive a single high power load? I know you have the purpose-built 286 and 686 designs, but can multiple modulus 86's be used to achieve the same thing?
Not if you want good performance.
You could cobble multiple Modulus-86 in parallel with 0.1-0.22 Ω ballast resistors, but for the best performance, you need the paralleled output section within a global feedback loop, which is not possible with multiple MOD86es in parallel.
And forget bridging. To get good performance there, you'll want to turn the amp into a fully differential amp, which also is not possible with multiple Modulus-86es. Never mind that to reproduce a 6xLM3886 Modulus-86 bridge/parallel amp would cost quite a bit more than a single Modulus-686.
The MOD86 is about $125/channel for PCB + parts. 6*125 = $750. Versus $499 for the MOD686 assembled module (or $350ish if you go the SMD-prepopulated + mounting brackets + parts route).
Do note that assembling the Modulus-686 from the SMD-prepopulated boards requires quite a bit of patience and attention to detail. You really want to get all the LM3886es and the PCB mounted to the heat sink before you solder the LM3886es. Otherwise, you're guaranteed that the LM3886es won't line up properly with their mounting holes. De-soldering the ICs is nearly impossible without wrecking the PCB, so don't count on that.
Also, have a look here: Taming the LM3886: Output Power – Neurochrome
Tom
Hi,
I want to make a Modulus 86 stereo amp.
Any of you who have compared Toms linear Power 86 to Connex SMPS300RE ?
I want to make a Modulus 86 stereo amp.
Any of you who have compared Toms linear Power 86 to Connex SMPS300RE ?
I have. Get the ±30 V version of the SMPS300RE. The amp performs equally well on either supply.
One drawback of the SMPS300RE is that Connex sometimes cheaps out on the connectors, so you may have to replace the mains input connector. The cheap blue Chinese ones they sometimes use don't hold up to much of anything, and I've had several of them crack when I tightened the screws.
Also note that the SMPS300RE is protected against over-temperature on the transformer only. I've had a couple that died during sine wave testing as the output diodes overheated and fried. I've never had a failure during music reproduction, though. The failures were my fault, really. I overloaded the supplies and they didn't like that so much. That's not the supplies' fault, but it would have been nice with an over-temp sensor on the output diodes as well.
In the cases where the supplies failed, they failed gracefully. The output voltage just dropped to zero and that was it. Swap the diodes and the supplies were happy again.
Tom
I made a dual-mono Modulus-86 a while back. I already had a SMPS300RE laying around from a previous project so I figured it would be a cool idea to buy another and go nuts.
Attachments
Good eye! The BZ4309 is really nice to work with and built surprisingly well.
I love this amp. So much so that I recently ordered a pair of Guardian-86's. I've been having way too much fun building different DIY tube preamps that I figured it would be best to protect the amp in case something goes wrong.
Supposedly you get better channel separation, I dunno. I've never built a dual mono amp before so that was my main inspiration. The power supplies also generate very, very little heat. I'm hoping that will add to their longevity as they each have an easy workload.
In many builds you would get better channel separation with a dual-mono build. That's because you eliminate any interaction through the power supply by building dual mono.
For the Modulus series, there's no real benefit to the dual-mono configuration - at least not in terms of amplifier specs - as the power supply rejection ratio of the amp is incredibly high. However, as you point out, you do likely get a benefit in terms of longevity as each power supply only sees half the load.
Either way. The supplies are cheap and light and your build is well executed, so why not... 🙂
Tom
For the Modulus series, there's no real benefit to the dual-mono configuration - at least not in terms of amplifier specs - as the power supply rejection ratio of the amp is incredibly high. However, as you point out, you do likely get a benefit in terms of longevity as each power supply only sees half the load.
Either way. The supplies are cheap and light and your build is well executed, so why not... 🙂
Tom
