So you have nothing to counter the evidence from JBL themselves on how they develop technology. Until you do I maintain that getting 'reviewers' to help you develop products is flawed unless it is your sales tactic.
I am obligated to maintain confidentiality...
I faithfully present my views and practices when possible, you can have your own opinion, it really does not matter to me.
I do not rely on audio development for a living. Actually I have been retired from work for more than 10 years. I just got fed up with spending money on audio equipment out there on the market. This is the only reason why I spend time and money on development in this area. Yes, lots of friends think I am crazy to to this. Here is where all the crazy people are...
I faithfully present my views and practices when possible, you can have your own opinion, it really does not matter to me.
I do not rely on audio development for a living. Actually I have been retired from work for more than 10 years. I just got fed up with spending money on audio equipment out there on the market. This is the only reason why I spend time and money on development in this area. Yes, lots of friends think I am crazy to to this. Here is where all the crazy people are...
Bill, I've used DBLTs to design stuff for nearly 2 decades.
While a DBLT series would be used to confirm approval of the final product, an engineer would only use a DBLT to make difficult & subtle decisions.
A true golden pinnae (someone who has been proven reliable in a LOT of DBLTs) can often pick stuff up in seconds so much of the design work could be sighted. But even the best ears in the business are as prejudiced and pig headed as the most raving lunatic Golden Pinnae reviewer. There are ears whose opinion I respect blind that I would never consider sighted (including myself
).When you do a lot of DBLTs, you find out all sorts of stuff .. some of which I've pontificated on this forum in various places. eg practically ALL HiFi reviewers are deaf (with a VERY small number of exceptions).
ALL self declared Golden Pinnae are deaf. No exceptions
I could go on but that would further hijack Tom's thread. Most of the tests were of speakers but there were important tests on amps, EVIL digital and other stuff too.
I don't think most people realise how expensive it is to do proper DBLTs. A DBLT is a measurement and your instrument, your listening panel, has an accuracy that has to be checked & calibrated regularly.
But to come back to amps, I posted something something on another thread on what features would result if DBLTs were the MOST important tests conducted on amps. It will be a long time before amp makers & designers are brave enough to do this.
Last edited:
Your blog clearly indicates you are developing a product to sell, so this is not a DIY for home use effort. But everything you post suggests you are working on something that sounds 'nice' rather than looking for fidelity.
FWIW I do not believe that I could tell a Mod-86 from another 3886 based amp in blind testing. But I do know that KNOWING I have the best objective perfomance I can currently afford will enhance my listening pleasure and stop me worrying about power amps for another decade. I accept my biased brain and use that to my advantage.
You haven't seen my above mentioned list of features.
They are nothing to do with stuff hand carved from Unobtainium by Virgins.
But the list is shocking to both the Golden Pinnae & Objectivist brigades
Contrary to popular myth, there HAVE been DBLTs on amp 'features' that consistently result in better sound. But these features are too heretical for common consumption ... or even discussion.
But if you want to incorporate stuff that actually results in 'better sound' ...
And no. None of the amp designs on this forum .. or indeed any commercial amp .. incorporate the most important of these features.
Fidelity should sound nice. If not, why would we want to listen to music.
Yes, eventually I would want for people to enjoy what performers offer. But if you care to look at the time extension, you will find that there is no intention to release to production that is not satisfactory and something I would pay to have myself. Show me who else can go on for such a long time without releasing a product?
The main issue is integration. For example, an exporter wanted us to take the amplifier out of the active speaker and make a stand alone amp. There is now way this will happen unless there is another stand alone speaker to provide the best integration. As a matter of fact, a new driver is being developed for that purpose. It is difficult for some factories to handle this because they need the revenue from sales.
You need to look up the meaning of Fidelity. Boutique audio where egos are stroked and things are shiny might need to, but High Fidelity means accurate. High End means expensive and often fatally flawed.
If you are designing an active speaker, does beg the question what you are doing on this thread. Are you planning to license the Mod-86 or taking ideas for yourself? I think some disclosure is needed.
If you are designing an active speaker, does beg the question what you are doing on this thread. Are you planning to license the Mod-86 or taking ideas for yourself? I think some disclosure is needed.
I am observing how the modulus will progress. Currently there are many criteria that it does not meet compared with the topology now used in the active speaker. If those get addressed, then there is a possibility to license it.
MyRef was also considered, but did not meet the criteria either.
Accurate should sound good. The problem is that each part needs to be designed to certain criteria for the total system to be accurate. This is where most designers cannot deal with alone. If you look at the MyRef thread, you will find that it does address some of the issues I consider important, but not all. I have not seen any way to change it so that it does. The Modulus only focuses on distortion, although I had asked for other data, it was never provided. So it is still a mystery to me until I can get the time to put it together and see the tests done myself. So we shall see.
Ted Jordan had an interesting say " when all you have is a hammer, then everything looks like a nail" (hope I got that right).
MyRef was also considered, but did not meet the criteria either.
Accurate should sound good. The problem is that each part needs to be designed to certain criteria for the total system to be accurate. This is where most designers cannot deal with alone. If you look at the MyRef thread, you will find that it does address some of the issues I consider important, but not all. I have not seen any way to change it so that it does. The Modulus only focuses on distortion, although I had asked for other data, it was never provided. So it is still a mystery to me until I can get the time to put it together and see the tests done myself. So we shall see.
Ted Jordan had an interesting say " when all you have is a hammer, then everything looks like a nail" (hope I got that right).
For that reason in a bridged setup, I would use caps from + to - rails. There is minimal current flowing through the ground terminal in bridged setup; ground is mainly for reference and housekeeping in this situation, so you don't need large reservoir caps connected to ground.
Edit: it might be even considered a mistake to have reservoir caps between the rails and ground in bridged mode, since this halves the capacitance and therefore a cap of the same value between the rails is twice as effective (albeit at the cost of a higher voltage rating required).
Last edited:
Many people say that, but it first appeared in the early 60s Law of the instrument - Wikipedia, the free encyclopedia
Member
Joined 2006
Thank you.
So I hope you don't mind if I bring the thread back to a technical discussion of your design.
Every now and then the thread seems to meander off trail. Threads have been known to do that. It's all good. Technical questions are always welcome. Just be prepared for a technical answer...
Bridging two amp channels is a great way to quadruple the output power. The only drawbacks I can see are:
- Twice the BOM cost.
- Takes up more room in the chassis.
- Each amp half drives half the load impedance, hence, will need high current output.
The Parallel-86 was designed for a couple of reasons, including:
- Customer requests for a parallel amp.
- Customer needs for higher output power.
- To be able to drive lower load impedances and address 3) above.
I expect to have a more complete set of measurements and specs for the Parallel-86 by next week.
Unless you invent a material with superconductive properties at room temperature, there is no way around Ohm's Law. However, it is possible to keep the "ground pollution" from corrupting the amplifier output. All one has to do is to connect the reference/signal ground to the power ground with a low-impedance path right at the output connector. This is one of the "layout tricks" I employ in the Modulus-86 and Parallel-86 and this is one of the reasons the circuit performs so well.
There are a few DACs out there that can match the performance of the MOD86. The ESS9018 springs to mind. The CS4398 would work as well if memory serves. Designing a DAC that meets the data sheet spec of the DAC chip itself is a challenge and many get it wrong or end up degrading the performance of the DAC by choosing a lower cost output driver to lower system cost.
I have to set up my measurements to ensure that at the max output power, the signal lands right at the sweet spot (top of the input range) for the APx525 to measure the THD. So yes. I'm at the limit of the AP and maybe a bit beyond.
~Tom
Last edited:
That's my point exactly. Most listening tests - and I would go as far as saying all DIY listening tests - are sighted tests. As such, they are fundamentally flawed and should not be used to make significant design choices.
That said, I always listen to my amps before committing fully to a design. That's just a fundamental sanity check. It's a way for me to enjoy the fruits of my work. I don't use listening tests to make design choices. For the same reasons as I wouldn't use a known flaky instrument to measure circuit parameters in precision circuits.
~Tom
Last edited:
I'm curious what those criteria would be and what the scientific reasoning behind them is.
~Tom
Thanks Bill.
You are the first person on this forum who has expressed interest on stuff that DBLTs have shown to result in better sound.
I'm travelling with expensive internet at the moment but will search after Easter. IIRC, I posted in either Self's or Cordell's huge threads on their books.
BTW, I think Soong uses crude Current Drive in his active speakers.
Last edited:
Lots I have already mentioned throughout the thread and my communication with you. But since you all just focus on "disagreements", it hardly makes sense to go over it again.
Please review what I said about input impedance, amplifier current and phase, damping. And without losing the low distortion measured here.
Last edited: