Twisted pairs work for differential signaling if the common-mode rejection is sufficiently high. For single-ended signaling, it just looks pretty...
Now, keeping the wires tightly coupled is usually a good idea, however. That has more to do with loop area and coupling inductance, however.
~Tom
Now, keeping the wires tightly coupled is usually a good idea, however. That has more to do with loop area and coupling inductance, however.
~Tom
I may be suffering from sleep deprivation induced blindness... Which connectors are those? Do you happen to have the part number handy? I looked at Neutrik's catalog this morning and couldn't find any with built-in filters.
~Tom
The BOM, full schematics, and link to the relevant Mouser project are supplied with the boards. Also included is a design document which explains the details of operation of the circuit, the various options on the board, and how to assemble and hook up the board for operation. I'm currently a bit behind on the design doc but expect to have it ready over the weekend.
All the documentation is sent out to the email address used with Paypal, so make sure it can handle a couple of MB worth of .pdf.
~Tom
All the documentation is sent out to the email address used with Paypal, so make sure it can handle a couple of MB worth of .pdf.
~Tom
Ordering the BOM from Mouser is easy. Tom gave me a link and the BOM is already there just click pay. It cost me $240.00AUD for 4x parts kit so $60.00AUD each. This would be about $220USD for a stereo pair including PCB's
For common mode filtering you need the connection to the chassis/shield. If the cable is feed through the chassis without a strong (or at least a feedthrough capacitive) connection to it at the point of entrance you allow the RF noise carried on the outside of the shield/gnd conductor to enter the enclosure and contaminating the GND/reference potential on the PCB).
One can design on-board filters that do work, of course, with that in mind -- just don't connect the shield to the local GND, only to the filter network, and connect the shield to chassis at the point of entrance.
Arh, OK. I wasn't going crazy. I was looking at the chassis connectors not the cable connectors.
~Tom
For those interested in shields and grounding, this is $15 well spent.
SHIELDS AND GROUNDS: SAFETY, POWER MAINS, STUDIO, CABLE AND EQUIPMENT, (special excerpt) The June 1995 issue of the Journal was a definitive and comprehensive collection of information on this important topic. The seven papers by Neil Muncy and other experts in the field have been reprinted into a convenient guide for designers and practitioners.
Special Journal Issues
SHIELDS AND GROUNDS: SAFETY, POWER MAINS, STUDIO, CABLE AND EQUIPMENT, (special excerpt) The June 1995 issue of the Journal was a definitive and comprehensive collection of information on this important topic. The seven papers by Neil Muncy and other experts in the field have been reprinted into a convenient guide for designers and practitioners.
Special Journal Issues
What are the recommendations for the power supply?
It's a real shame that the DIY community cannot look at the schematic of this project, it hinders the enjoyment from learning new things by dissecting circuits. If you want your design to evolve and be analyzed by word class engineers than sharing is a must.
It's a real shame that the DIY community cannot look at the schematic of this project, it hinders the enjoyment from learning new things by dissecting circuits. If you want your design to evolve and be analyzed by word class engineers than sharing is a must.
Agreed.
There appears to be no speaker protection circuitry but without a schematic it's hard to tell. What happens if the LM3886 output shorts to the V+ or V- supply rail?
Little bit of contradiction in the latest drift of the thread here, I think. An engineer familiar with composite amplifiers, whether world class or merely competent, will have no problem inferring the schematic from the topology diagram and running the simulations to identify the design point chosen within a couple dB. If, as a DIYer, one chooses to go the Design it Yourself route Tom's provided numerous references, quite a bit of guidance on how to get started, and extensive 3886 measurements and layout information over in the chipamps forum. All pro bono. This well beyond what most DIY vendors offer---even ESP doesn't offer as much layout insight.
Note that limiting sharing to topology information is normal in higher end class D builds and in vendor offerings. As Bruno, Eva, Choco, and others all regularly point out the prerequisite information's there if one wants to learn how to DIY it. If one prefers not to invest in personally moving up the learning curve there's a very good alternate option in buying the board and being happy. Both with the performance and compensating someone who seems to be a pretty damn honest businessman for his contributions to the community.
Note that limiting sharing to topology information is normal in higher end class D builds and in vendor offerings. As Bruno, Eva, Choco, and others all regularly point out the prerequisite information's there if one wants to learn how to DIY it. If one prefers not to invest in personally moving up the learning curve there's a very good alternate option in buying the board and being happy. Both with the performance and compensating someone who seems to be a pretty damn honest businessman for his contributions to the community.
The topology diagram seems quite clear in not showing protection circuits beyond the 3886's internal current limiting and thermal shutdown---I'm sure the man himself will confirm but it seems vanishingly unlikely someone would do a board with these features and then forget to include them in the documentation. This makes sense; for a board oriented towards bridging and paralleling you'd probably want to include that circuitry once at the combined output rather than paying to repeat it in each board's build.
agree 100%
Toms design in itself and expertise make the PCB and BOM interesting to me, as I am essentially lazy. That being said I have been considering attempting a composite design of my own, amateurish as it would likely be. Kindly Tom has provided a large amount of info in that regard, and I may well try it myself.
I am essentially undecided. I haven't built my Audiosector 4780 PCB yet however, as time has been sparse. Once I get that completed then ill have more confidence in attempting another design, perhaps my own.
Even the most open and honest businessman and designer isn't going to give their design away
Common sense really.
Toms design in itself and expertise make the PCB and BOM interesting to me, as I am essentially lazy. That being said I have been considering attempting a composite design of my own, amateurish as it would likely be. Kindly Tom has provided a large amount of info in that regard, and I may well try it myself.
I am essentially undecided. I haven't built my Audiosector 4780 PCB yet however, as time has been sparse. Once I get that completed then ill have more confidence in attempting another design, perhaps my own.
Even the most open and honest businessman and designer isn't going to give their design away
Common sense really.
Everytime someone gives their design someone else copies it and sells pcbs. He's trying to make money. Let him do it without your contribution if your ethical philosophy points in another direction. Gimme a break with the pressure to give up the schematics. I sold the Lighter Note and in those 5 years shared it with 2 people. It still works and sounds great and is not revolutionary. If I'd shared it it would have immediately lost value as a kit. I still help people build their own versions just as recently as yesterday. I bet Tom has similar feelings.
The people who demand schematics but wont make a purchase almost 100% of the time dont know what to do with schematics to start with. So what is the benefit of posting them? He doesn't need your help or approval. He already has proven its performance. Again no benefit to posting.
The people who demand schematics but wont make a purchase almost 100% of the time dont know what to do with schematics to start with. So what is the benefit of posting them? He doesn't need your help or approval. He already has proven its performance. Again no benefit to posting.
What? You say businesses have to generate enough profit to make it worth staying in business? Who knew?
A key skill in composite amp design is being sufficiently comfortable with phase and gain margin to work with multi-loop topologies where the notion of loop gain is not well defined. Big jump from populating a single stage board without servo to parameter sweeping rate of closure across component tolerances, load variations, and closed and open loop (clipping or SPiKe) conditions. If you get it wrong the amp will eventually encounter an operating point where it goes unstable. I've gotten some of my composite boards taped out only to abort the PCB run at the last minute because I realized in final design review I'd overlooked a particular corner case in sim and needed to revise the circuit.
The problem here is that the LM3886 has been used for quite some years, and implemented in dozens of DIY amp's. It would be nice to see what exactly Tom's implementation is going to offer. As for Duplication, it's only a matter of one person spending $60 and drawing out the circuit. But why waste the time?
IMHO over 90% of DIY-audio members are loyal enough to support the Vendor. If not SHAME SHAME...
IMHO over 90% of DIY-audio members are loyal enough to support the Vendor. If not SHAME SHAME...
Last edited:
Morrison's book on the topic is a good read as well:
Morrison, R: Grounding and Shielding: Circuits and Interference.
@Andrew: The forum admins are probably not the right people to ask. You need to ask AES if they will allow a copy of their copyrighted material to be distributed free of charge to DIY Audio members for educational purposes (fair use).
~Tom
Morrison, R: Grounding and Shielding: Circuits and Interference.
@Andrew: The forum admins are probably not the right people to ask. You need to ask AES if they will allow a copy of their copyrighted material to be distributed free of charge to DIY Audio members for educational purposes (fair use).
~Tom
Yup, clones and fakes are occupational hazards of any business. Pretty sure getting into the various points of order for intellectual property law and protection is off topic but it's good some protections exist. As udailey touches on, there's no reason to make copying free.
Post 1 of this thread and its various data followups are objectively far more specific as to what the amp has to offer than the great majority of amps---others which are as well characterized are few and far between. The subjective side I imagine will take time as boards have to be sold, folks have to build them up, and so on. I'll also predict the discussion will move over to, gee, a good amp can reveal a poor speaker just like a good speaker reveals a poor recording. With luck there'll be some folks with measurement mics who can provide A/B data of their systems for clarity.