Thank you, yes, that's definitely an option. I'm thinking I might try to emulate the LEDs in the front panel of my NAD3020A, where just the hemispherical dome of the 3mm LED protrudes above the front face - looks very neat and minimalist.
An externally hosted image should be here but it was not working when we last tested it.
This would need the front panel to be drilled to two different diameters - 3mm for the outer hole, and a fraction more, say 3.2mm, for the base of the LED. Like this maybe:
An externally hosted image should be here but it was not working when we last tested it.
I think a fixed length stand-off on the PCB would support the LED better, so the 3.2mm hole would have to suit the diameter of that stand-off. This one would need a 6mm hole.
It depends somewhat on the whether the front panel provider can drill those two different diameters accurately.
This is what I've ordered:
I shuffled a few components around so that I could make room for the header pins to allow the LEDs to go off-board. I decided there probably wouldn't be room to mount the LED control board on the rear of the front panel, so I made a much smaller mounting board, also with header pins so that it can easily connect to the control board.
This is much more compact, so I'll have much more freedom to mount it behind the front panel. The LEDs are at 10mm pitch, just like my NAD amp. I also included a space for the DLCP 'Power' LED.
The idea would be to mount the LEDs in fairly tall spacers so that they stand a fixed distance above the PCB, and then mount the PCB at the right offset from the back face of the front panel, so that the LEDs sit just proud of the front face.
An externally hosted image should be here but it was not working when we last tested it.
I shuffled a few components around so that I could make room for the header pins to allow the LEDs to go off-board. I decided there probably wouldn't be room to mount the LED control board on the rear of the front panel, so I made a much smaller mounting board, also with header pins so that it can easily connect to the control board.
An externally hosted image should be here but it was not working when we last tested it.
This is much more compact, so I'll have much more freedom to mount it behind the front panel. The LEDs are at 10mm pitch, just like my NAD amp. I also included a space for the DLCP 'Power' LED.
The idea would be to mount the LEDs in fairly tall spacers so that they stand a fixed distance above the PCB, and then mount the PCB at the right offset from the back face of the front panel, so that the LEDs sit just proud of the front face.
While I wait for my boards to arrive I've been experimenting with the Amp Ready signal from a single amplifier module.
I mocked up this circuit:
like this:
Difficult to see in that jumble of wires, but it does correspond to the schematic.
VDR is +15.6V
R1 is 1400 Ohms, the LED is 2V, 10mA.
R2 is 20k
The transistor is one of these.
When the amp powers on, the amp ready pin goes to approx 1.9V briefly, then drops to -0.7V. When it drops to -0.7V, the LED glows brightly, as it should.
But while the RDY pin is briefly at 1.9V, the LED glows dimly - enough to be ambiguous as to whether the amp is ready or not - and I can't figure this out.
I assumed the transistor would either be on or off, so either no current would flow through the transistor, or it all would. So why, when the base is at 1.9V, does the LED get enough current to glow dimly?
I mocked up this circuit:
like this:
Difficult to see in that jumble of wires, but it does correspond to the schematic.
VDR is +15.6V
R1 is 1400 Ohms, the LED is 2V, 10mA.
R2 is 20k
The transistor is one of these.
When the amp powers on, the amp ready pin goes to approx 1.9V briefly, then drops to -0.7V. When it drops to -0.7V, the LED glows brightly, as it should.
But while the RDY pin is briefly at 1.9V, the LED glows dimly - enough to be ambiguous as to whether the amp is ready or not - and I can't figure this out.
I assumed the transistor would either be on or off, so either no current would flow through the transistor, or it all would. So why, when the base is at 1.9V, does the LED get enough current to glow dimly?
Perhaps the transistor is too low in beta (current gain), HV types can have relatively low beta.
Use a 2N3904 or 2N2222A instead. If you have to use the HV transistor, try reducing the
base resistor from 10k to 2k or less.
Transistors can act as on-off switches, but only if used in certain ways. Otherwise they will act
as quasi-linear amplifiers, with an output related to the magnitude of the input, rather than its
presence or absence.
Last edited:
It's always best to breadboard a circuit with the exact type of components intended to be used,
before committing the circuit to a pcb. That's why you see traces cut, wires patched on, and
components tacked onto prototype boards, because some changes had to be made to get the
circuit to work right.
Last edited:
Indeed. I'm hoping that the basic board layout will work with whichever components turn out to be suitable. If not, then it's only another week and a few pounds extra to have 10 more PCBs made with a revised layout. It wouldn't be much of a hobby if everything worked first time and I didn't have to learn from my mistakes.
I've been building circuits for a long time, and haven't had a pcb work perfectly on the first try yet.
Don't know anyone else who has, either.
As suggested, a 2N3904 fixed it (one of these). The LED now stays off until the signal drops to -0.7V. Marvellous.
Now I need to figure out a way to test the clipping and current limiting circuits, without risking damage to the amplifier. I guess for clipping I need to find a way to supply a source with a voltage that's too high. Not sure about current limiting though - not sure what conditions would prompt the amp to go into current limiting, or whether it would be wise to deliberately create those conditions.
In fact, the current limiting circuit is the same as the Amp Ready circuit (on the adapter board at least - different logic for the front panel LED though), so if it works for the amp ready signal it should work for the current limiting signal. The clipping signal differs in that it's not internally pulled up - my board does the pulling up - and therefore may need the diodes to keep the transistor's emitter a little bit above GND.
Now I need to figure out a way to test the clipping and current limiting circuits, without risking damage to the amplifier. I guess for clipping I need to find a way to supply a source with a voltage that's too high. Not sure about current limiting though - not sure what conditions would prompt the amp to go into current limiting, or whether it would be wise to deliberately create those conditions.
In fact, the current limiting circuit is the same as the Amp Ready circuit (on the adapter board at least - different logic for the front panel LED though), so if it works for the amp ready signal it should work for the current limiting signal. The clipping signal differs in that it's not internally pulled up - my board does the pulling up - and therefore may need the diodes to keep the transistor's emitter a little bit above GND.
My Dissipante case arrived this week, so I've been doing some trial assembly. I bought the optional baseplate, which has a 10mm x 10mm grid of 4mm holes for mounting components without having to drill the chassis bottom plate. I asked for the front and back panels to be retained so that they can be customised later once I settle on the layouts. In the meantime the case holds together quite well with just the sides and the baseplate. In fact the sides are the heatsinks themselves - quite significant chunks of black anodised aluminium with a rather nice wavy profile to the fins. The whole chassis has a nice quality feel.
Here's a photo of most of the components assembled onto the baseplate. Just one amplifier module so far, as a sort of 'proof of concept'.
To mount the amplifier module I drilled and tapped some holes into the heatsink so that I could fit some M3 studs to replace the M3 countersunk screws on the UcD400OEM modules. Lousy photo, but you'll get the idea:
And here's an amplifier module attached to the heatsink. On final assembly I'll use some thermal paste to ensure a good contact.
This is the amplifier module attached to one of my early adapter boards - this is why I wanted the adapters to sit at 90 degrees to the amplifier. Unfortunately one of the key dimensions I was working to was 9mm out, so the adapter mounting holes don't line up with the baseplate holes, but I need another iteration of the adapter board anyway, so I'll fix that then.
I made some mounting plates for the DLCP and the SMPD SLCP so that they can be mounted on the 10mm x 10mm baseplate grid.
Here's a shot of the arrangement of my various PCBs. It shows the power supply splitter (with only three of the six positions filled with sockets so far), the USB relay board, and the 'Paper Circuit Board' representing the board that will aggregate the amp ready, clipping and current limiting signals into LEDs for the front panel (the real thing should arrive this week).
After planning, mocking up, designing PCBs etc for so long, I realised today that I could finally make a few more connections and power it all up. And it seems to work. I haven't put any audio through the amplifier module yet, but at least the DLCP starts up, and is accessible through the USB port.
I finally found a decent crimper (the Engineer PA-09) for the tiny 'Dupont' crimps used in these header pin sockets, and now the task of making all the cables that daisy chain the 'DC Error' and 'Amp On' signals back to the SMPS, and those which take the 'Amp Ready', 'Current Limiting' and 'Clipping' signals back to my aggregator board, doesn't seem so daunting anymore.
Here's a photo of most of the components assembled onto the baseplate. Just one amplifier module so far, as a sort of 'proof of concept'.
To mount the amplifier module I drilled and tapped some holes into the heatsink so that I could fit some M3 studs to replace the M3 countersunk screws on the UcD400OEM modules. Lousy photo, but you'll get the idea:
And here's an amplifier module attached to the heatsink. On final assembly I'll use some thermal paste to ensure a good contact.
This is the amplifier module attached to one of my early adapter boards - this is why I wanted the adapters to sit at 90 degrees to the amplifier. Unfortunately one of the key dimensions I was working to was 9mm out, so the adapter mounting holes don't line up with the baseplate holes, but I need another iteration of the adapter board anyway, so I'll fix that then.
I made some mounting plates for the DLCP and the SMPD SLCP so that they can be mounted on the 10mm x 10mm baseplate grid.
Here's a shot of the arrangement of my various PCBs. It shows the power supply splitter (with only three of the six positions filled with sockets so far), the USB relay board, and the 'Paper Circuit Board' representing the board that will aggregate the amp ready, clipping and current limiting signals into LEDs for the front panel (the real thing should arrive this week).
After planning, mocking up, designing PCBs etc for so long, I realised today that I could finally make a few more connections and power it all up. And it seems to work. I haven't put any audio through the amplifier module yet, but at least the DLCP starts up, and is accessible through the USB port.
I finally found a decent crimper (the Engineer PA-09) for the tiny 'Dupont' crimps used in these header pin sockets, and now the task of making all the cables that daisy chain the 'DC Error' and 'Amp On' signals back to the SMPS, and those which take the 'Amp Ready', 'Current Limiting' and 'Clipping' signals back to my aggregator board, doesn't seem so daunting anymore.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.