KSTR said:No, not overkill. Just thing of the ballpark 120V (60V for 120V mains) transient that an input has to withstand when you connect non-safety-grounded gear to safety-grounded. Or the typical human ESD pulse (rubber soles + synthetic carpet + dry climate).
My criterion is more stringent, any input (or other user accesible connection) should withstand the ~4kV surge pulse from a piezo oven lighter, directly fed into it. A piezo oven lighter with a horn or dipole antenna also is an excellent test tool to check if the microcontrollers/etc in your AVR/CDP or what have you is competently designed, that is, wether the whole device is protected against ESD events or not
EDIT resource: http://www.emcesd.com/. look a the tidbits section
EDIT2: and I'd sacrifice a toe (with nail polish, for Walter Sobchak) for his HP54845A...
- Klaus
I'm very well aware of ESD and all the protection methods.
But let me understand exactly: so you design your high end audio equipment with spark gaps on the PCB and ESD diodes at the inputs?
Have you seen any commercial high end audio equipment designed and built with such features?
I don't mean here studio equipment, that doesn't qualify in my view as high end. And all the today's highly integrated digital controllers and logic have all the ESD protections included on the chips.
syn08 said:
This whole connector interface pcb is an engineering solution for a non-issue, as is this concern for RFI in audio freq reproduction.
If you design a circuit (that in reality only needs to reproduce a reasonable audio bandwidth) to be flat from DC to light and then have problems when someone dials Mom on ther cell phone, where is the problem?
I know, I don't undderstand the big picture....
Cheers,
me
I think this is an exercise in excess and I will explain later, but for making disks a board shop can route rings that can be snapped out after assembly. Its very common practice and wastes little material.
Getting the vias will require a shop (most DIY'ers don't have access to the plate through process) so the routing is a simple addition. The surface mount caps will lower the impedance to the external enclosure which will reduce the RFI that gets in.
However insulated washers may be enough to create a good rf coupling when used with a 100 pF cap to the chassis. The goal is to keep the RF on the outside of the chassis and the audio inside.
It has worked well in high RF environments. Has anyone measured the capacitance across the washers?
A bigger issue is the size of the holes in the chassis, they all need to be small and short.
The NEC is an inappropriate standard for looking at an electronic component. UL has much more appropriate requirements. The fusing on the line input should be adequate to remove power inside the box if there is a short between line and the chassis. The 20A circuit breaker requirement (500A w/o failure actually) is for the wiring in the building, not for the device plugged into the outlet. A hipot tester with the ground bond test should be fine for confirming that its all safe.
Getting the vias will require a shop (most DIY'ers don't have access to the plate through process) so the routing is a simple addition. The surface mount caps will lower the impedance to the external enclosure which will reduce the RFI that gets in.
However insulated washers may be enough to create a good rf coupling when used with a 100 pF cap to the chassis. The goal is to keep the RF on the outside of the chassis and the audio inside.
It has worked well in high RF environments. Has anyone measured the capacitance across the washers?
A bigger issue is the size of the holes in the chassis, they all need to be small and short.
The NEC is an inappropriate standard for looking at an electronic component. UL has much more appropriate requirements. The fusing on the line input should be adequate to remove power inside the box if there is a short between line and the chassis. The 20A circuit breaker requirement (500A w/o failure actually) is for the wiring in the building, not for the device plugged into the outlet. A hipot tester with the ground bond test should be fine for confirming that its all safe.
Magura said:
No idea, but make sure you leave at least 7mm between the rings, for the end-mill.
Magura
Come to think of it, why must the little board be round? Just a square 0.0*0.8 inch with a center hole should do. Mount it in a Neutrik cup or by itself on a panel cutout with two screws.
Jan Didden
janneman said:
Sorry, missed those. Huge indeed; took me 10 mins to download. Not too inviting
Anyway, I don't think these are the same we were talking about.
No, it is a different solution for the same problem + of possible ground loop problems. XLR sockets bolted to chassis shield all wires, chokes (wires with ferrite beads on them) on each of symmetrical wires, then a PCB clad shields sockets from the rest.
syn08 said:
Could you elaborate on this pin 3 connection. I am interested to understand why. The basic idea there should be to divert high currents to avoid common impedance coupling at input or in power supply loop
As for the leakage of RF at the input connector, if the signal ground is connected to the back to back diodes which are then connected to the case via a ferrite choke, don't you think that the RF on the shield will leak less to the signal ground and flow to the case.
JPV
JPV said:
I have never seen the LME49810 internal schematic, so I can only speculate as of why. But if you want to get a grasp on what's up, get a LME49810 standard amp, connect pin 3 to ground through a small resistor (1 ohm will do) and connect a scope across. You will quickly decide you don't want that **** injected into your signal ground.
AND what does this have to do with the BLOWTORCH? The 'mice' play while the 'cat' is away?
Seriously, I think that much good input has been made about RFI proofing, even though it is somewhat 'over the top', and relatively pointless in an audio product.
We are not EMP proofing against enemy destruction of our communications. By then, we might as well say goodbye to each other and hope for salvation.
If we have a microwave tower aimed directly at us at that level, we will better move anyway, before we get cancer or something. I am serious about this. Still, it is an excellent intellectual exercise it what has been done, when it is necessary.
Seriously, I think that much good input has been made about RFI proofing, even though it is somewhat 'over the top', and relatively pointless in an audio product.
We are not EMP proofing against enemy destruction of our communications. By then, we might as well say goodbye to each other and hope for salvation.
If we have a microwave tower aimed directly at us at that level, we will better move anyway, before we get cancer or something. I am serious about this. Still, it is an excellent intellectual exercise it what has been done, when it is necessary.
PMA, it would seem to me that IF the input waveform, itself has a high level RF spectrum, then what good is it to reduce RF input from incidental radiation?
It is true that RF does exist in the air, and that it gets picked up by the inputs of audio electronics. In a typical worst case, we get the dreaded 'TV Sync buzz' due to the rectification of the TV RF signal by the input device, usually a bipolar transistor.
Fets and tubes are less susceptable, AND the 50-1K input resistor is there to reduce this RF, when used with an input bypass cap of 50-200pF. However this bypass cap is necessary for another reason. It is there to keep a low input impedance at the input at RF frequencies in order to keep the unit from oscillating, due to added phase shift.
It is true that RF does exist in the air, and that it gets picked up by the inputs of audio electronics. In a typical worst case, we get the dreaded 'TV Sync buzz' due to the rectification of the TV RF signal by the input device, usually a bipolar transistor.
Fets and tubes are less susceptable, AND the 50-1K input resistor is there to reduce this RF, when used with an input bypass cap of 50-200pF. However this bypass cap is necessary for another reason. It is there to keep a low input impedance at the input at RF frequencies in order to keep the unit from oscillating, due to added phase shift.
- Status
- Not open for further replies.