Welcome aboard. Pardon the turbulence in the thread at the moment. Unfortunately, some people prefer to make everything about them. I hope you're able to see the posts in their proper context.
Yep. The Universal Buffer can work with the Purifi and Hypex amplifier modules.
I have designed custom buffers for the Hypex NC500 and Purifi 1ET400A for a commercial client in the past. These buffers were designed specifically for his needs and included features not found in the Universal Buffer.
If you need me to design a buffer to meet your specific requirements, I'm more than willing to take that on as a consulting project. Just toss me an email (use the Contact Us form on my website) and we'll get the specs and pricing figured out.
I have no plans for making a buffer available to the DIY community for use specifically with the NC500/1ET400A. I think that's a bit too narrow of a market to go after in a DIY sense. That's not to say that I can't make a small adapter board that would make it easier for you to interface my Universal Buffer to the NC500/1ET400A, however.
Tom
Thank you.
So is the math around the addition of two tones of different frequencies, as is done in an IMD test.
Tom
No, he is exactly right.
As for "The IMD degrades as the amp enters clipping. Any amp does that.":
Measured at 3 Volts output, with 30V supply. That means "enter clipping?"
60 Hz/7KHz 4:1. IMD with 60 Hz? Now that's what I call fast.
And then with a 10K Load.
Everybody can do weight lifting with a Zeppelin.
But for channel separation, the load is 600 Ohm again, because it helps.
A buffer for 600 Ohm load? That is a buffer? And with important specs measured at 10K?
That's the same shitty specmanship to impress newbies as the potato semi cmos "buffers"
that could not even drive their own input.
NahI see. So you are working on reverse-engineering it?
Not this one, I promise. But I had to have a look at the simplified schematic and the board to see if sonic bliss' arguments held any merits. I've been reading D. Self small signal book a lot recently and working with a design using opamps at low impedance so the design appeared very familiar. I must say btw that this lm4562 datasheet is a tiny bit confusing at first sight as the graphs you quoted omit the "rms" for the output voltage. More recent datasheets clear that up.
Having used ncore 400s I can safely say one will never turn the wick up. I never used my khozmo preamp much past -30db, driving one to max output is humerous beyond belief.
I run Tom's mod86p rigged at 20db gain, even with a unity gain buffer pre, a balanced 4v dac output drives them to way beyond comfort level.
26db gain is way too much for modern sources
I run Tom's mod86p rigged at 20db gain, even with a unity gain buffer pre, a balanced 4v dac output drives them to way beyond comfort level.
26db gain is way too much for modern sources
Excellent. Your analysis was spot on. I'm glad someone reads my website.
That's an excellent read, actually.
I agree with that. There's an insane number of graphs in the data sheet. Do we really need data both for ±15 V and ±17 V supply? #whatever
At least the information is there. Some manufacturers barely give you enough to design from (and in some cases not enough to even make a qualified decision as to whether to use the part).
Yeah. My RME ADI-2 DAC only very rarely makes it above -30 dB with my Modulus-686 (26 dB gain) driving KEF R700 in a comfortably sized living room.
Tom
I have absolutely no axe to grind here. I don't know Tom, and indeed live on a different continent. But his products make total sense. They knock spots off most commercial offerings, and are clearly designed by a guy who has a deep understanding of solid design principles, and a raft of expensive test gear he clearly knows how to use.
But I am an electronics engineer (and laser physicist for what its worth), and I am confused where you are coming from Sonic Bliss.
If you don't like Tom's buffer - no one is forcing you to buy it. Buy something else you think will float your boat.
And as far as your lack of ability to understand spec sheets and basic circuit design - buy a book, go on a course, or don't. I honestly don't give a damn. Just stop wasting everyone's time.
How strange and disappointing that this sort of nonsense goes on.
Tom is a top guy, his amps are unbelievable and a veritable bargain in the hifi world. His 8x2 preampis equally as stunning and I don't doubt the buffer would be anything less than the same.
Tom, all this talk about Purifi and balanced inputs has certainly had some legs here since I last checked in. My interest is neither of those things 'cause my sources are about 1.5vrms output (max.) and the amplifiers are single ended 5kR input imdedance and clip at 4-5vrms, hence the need for some gain in the buffer (12dB seems to work well). When you get to it, I would really appreciate some single ended measurements at somewhere near those levels. I know I've asked before but hopefully it has not been lost in all this other noise lately.
Anthony
Anthony
No you should be ashamed of your self for not understanding the facts, and falling for the smoke screen.
Ok so based on how Tom described the circuit layout, with a differential input, half the voltage is split between each rail, and from there fed into the LME49724.
Unless there's a single ended input. In this case one of the rails of the LM4562 needs to deal with the full 9.6v in order to drive the amp to full output (which happens to exceed the 8v IMD clipping point in the data sheet by a whopping 1.6v)
So based on how Tom described the circuit, the board will crap out 1.6 volts before the amp, with a single ended input. So this means no crazy beer drinking parties with the boys with this buffer, if your source is single ended. You need to keep it down. Might make the wife happy anyways.
But what about if you have a differential source? This means the LME49724 will need to be able to drive 9.6v. And he says it will be clean for 1 reason and 1 reason only. Because the datasheet just happens to be missing the IMD data for this particular opamp!
How convenient. No data on the datasheet means it must not matter right? WRONG. No data on the datasheet usually is a good indication that the performance is poor in that department. Hence the need for multi-tone IMD data of the finished product driving a 10V output into 2-4k from both a single ended source, and balanced source.
No kidding! I had not expected that either. I designed the Universal Buffer as exactly that ... a Universal Buffer. I intended it as a preamp puzzle piece, where 10-100 kΩ loads are common.
It wasn't until after launch when people started contacting me about its use with the Purifi and Hypex modules that I realized I could better position the buffer - you could say, make it even more universal - by characterizing its performance with a 4.4 kΩ load to model the input impedance of the Purifi and Hypex modules. If it will help more people buy the Buffer, I'm all for it. As I've mentioned numerous times already, I'm sure the Universal Buffer will show stellar performance.
So far, I have on my list:
- THD+N vs output level at various gains (I'll test all the values I list in the design doc).
- THD+N vs output level at various load impedances.
- Channel separation vs load impedance.
Higher up on my priority list is: Get the TCA HPA-1 and a product that has yet to be announced ready for the Florida Audio Expo. I need to design some signage and get a booth setup going. I fly out in two weeks, so this has top priority.
Tom
Why are you so worried about providing the IMD data? Why can't you provide all the same charts you have on your website for the 2v/100k load, for 10v/4.4k? You yourself state clearly on your website that the multi-tone IMD is the most important indicator of sound quality. So what makes you think the guys using this for a Purifi buffer have no interest in sound quality?
Thanks and good to know! I appreciate the reply. Have you mentioned this small adapter before? I’m trying to dig back and find the genesis of what I thought I read. Going off memory I think this person contacted you before you released your universal buffer and that’s where I may have gotten confused as I didn’t realize you just released it for purchase and the threads I was reading were a few months old.
In any case thanks again for reply!
I have not.
That could be. A few have contacted me regarding an adapter board. I have yet to actually design it. I was initially dismissive of the idea, but with the renewed interest and with some further development of the idea in my mind, I'm coming around on that. I think I can put together an adapter board that would be useful both for the DIYers and for OEMs.
The Universal Buffer opened for preorders in December and has been in stock for about a week now.
Oh, and for those looking for the bulk discount: The bulk/volume discount on the Universal Buffer has now been debugged and is operational. The price breaks are: QTY = 3, 5, and 10.
Tom