Bh dac q
- Thread starter xcortes
- Start date
John Swenson
New member
xcortes said:
It's right around 500 ohms. There is NO active output stage, the output from the DAC drives a simple final analog RC filter, which is what mostly determines the output impedance.
John S.
Grainger49
New member
John,
Sorry to bother you with this but I am planning to have 1.5M of unshielded cable to my Foreplay. Would you see any problem? I don't know the capacitance per foot for it.
Sorry to bother you with this but I am planning to have 1.5M of unshielded cable to my Foreplay. Would you see any problem? I don't know the capacitance per foot for it.
Output voltage is 2V with the filter incorporated into the DAC chip, and 1V with my favorite filter, John's filter that is implemented in the FPGA.
And Grainger, the same rules about cable length as always apply to the DACs connection to other gear. It depends in part upon the input impedance of what you run the DAC into and in part upon the capacitance per foot of the cable. A 500 ohm output (same as Beepre's single ended output) is pretty low and should work with pretty much anything you throw at it.
And Grainger, the same rules about cable length as always apply to the DACs connection to other gear. It depends in part upon the input impedance of what you run the DAC into and in part upon the capacitance per foot of the cable. A 500 ohm output (same as Beepre's single ended output) is pretty low and should work with pretty much anything you throw at it.
John Swenson
New member
That 1V output with my filter is a temporary thing, by the time it gets into production I'll tweak the output sevel back up to near 2V level.
It can't be right at 2V level, the filter has to have a slight attenuation to deal with files that have been boosted to absolute full scale, when you run this through an upsampling filter it can actually go to a level grater than the max, which causes clipping distortion, not good. So the filter has a slight attenuation to deal with this.
When I built this filter I accidentally specified 6db instead of 1 db. I'll get this fixed when I'm doing the firmware for the production board.
John S.
It can't be right at 2V level, the filter has to have a slight attenuation to deal with files that have been boosted to absolute full scale, when you run this through an upsampling filter it can actually go to a level grater than the max, which causes clipping distortion, not good. So the filter has a slight attenuation to deal with this.
When I built this filter I accidentally specified 6db instead of 1 db. I'll get this fixed when I'm doing the firmware for the production board.
John S.
The boards will be programmed and tested before we ship them. We will be working the details of the process out with John. The battery charging circuit which is constructed by the builder will also have documentation for a test procedure before connecting it to the expensive DAC board.
Grainger49
New member
Dan,
I had thought that the SMD components would be on the board and the through hole mount components would not. It sounds like that is wrong.
My existing DAC uses 2 - 12V SLA batteries. They are a +/- 12V supply when playing and in parallel to charge. That is one large wall wart charger!
I'm getting excited about this one.
I had thought that the SMD components would be on the board and the through hole mount components would not. It sounds like that is wrong.
My existing DAC uses 2 - 12V SLA batteries. They are a +/- 12V supply when playing and in parallel to charge. That is one large wall wart charger!
I'm getting excited about this one.
would be able to John Swenson
New member
There are two sets of firmware on the board: FPGA and XMOS processor.
The FPGA is the "brains" of the whole thing. It is the S/PDIF decoder, buffer controller, output format generator, adjustable oscillator interface, display controller and several others I can't think of at the moment. The firmware for the FPGA is stored in a special flash memory that is soldered to the board. In order to reprogram this flash memory there is a socket into which you can insert a cheap "eeprom". There is a switch on the board that takes the data from this eeprom and programs the FPGA flash. This chip in the socket is only needed when new firmware is being installed. From the factory there will be no chip in the socket. When new firmware is needed the Bottlehead gang will ship you a little chip, you put it in the socket and flip the switch. When it is done you can take out the chip.
Since there is a USB port I could have done firmware upgrades via USB, but there is a big proglem, you need a driver on the computer, which means I have to maintain drivers on every OS Bpttlehead users use, I REALLY did not want to do that! And every time Apple or Microsoft came up with a new version I would probably have to re-write the driver, not on my list of things I want to do forever. So yI decided on ship out something and plug it in, no software needed!
The XMOS processor also has firmware, but it is just for the USB interface, much lower probability of needing updating. As it is this is updated via USB, which unfortunately has the same problem of requiring a driver. It can theoretically be updated over the AUX port, so I might set things up so that any board that we send out that plugs into the aux port automatically updates the firmware for the processor on the main board.
John S.
The FPGA is the "brains" of the whole thing. It is the S/PDIF decoder, buffer controller, output format generator, adjustable oscillator interface, display controller and several others I can't think of at the moment. The firmware for the FPGA is stored in a special flash memory that is soldered to the board. In order to reprogram this flash memory there is a socket into which you can insert a cheap "eeprom". There is a switch on the board that takes the data from this eeprom and programs the FPGA flash. This chip in the socket is only needed when new firmware is being installed. From the factory there will be no chip in the socket. When new firmware is needed the Bottlehead gang will ship you a little chip, you put it in the socket and flip the switch. When it is done you can take out the chip.
Since there is a USB port I could have done firmware upgrades via USB, but there is a big proglem, you need a driver on the computer, which means I have to maintain drivers on every OS Bpttlehead users use, I REALLY did not want to do that! And every time Apple or Microsoft came up with a new version I would probably have to re-write the driver, not on my list of things I want to do forever. So yI decided on ship out something and plug it in, no software needed!
The XMOS processor also has firmware, but it is just for the USB interface, much lower probability of needing updating. As it is this is updated via USB, which unfortunately has the same problem of requiring a driver. It can theoretically be updated over the AUX port, so I might set things up so that any board that we send out that plugs into the aux port automatically updates the firmware for the processor on the main board.
John S.
mcandmar
New member
John Swenson said:
Oh please do, anything less than redbook levels would mean having to reconfigure everything i own to increase the gain. It wouldn't do the product any favors at all, i can only imagine the reaction on head-fi if you said it had an iPod level output.
Somebody mentioned the possibility of supporting multiple channels, is there any chance of fully balanced output via an add on card in the future?
John Swenson
New member
Doing an external balanced board COULD be done, but it will not be cheap. The best way to do it is with a separate DAC chip for each phase, I can't just add one, it would have to duplicate the: clock circuits, reclocker and have two DAC chips and associated components. Certainly doable, but not dirt cheap.
Whether it gets done I presume is determined by how many people want to pay for it!
John S.
Whether it gets done I presume is determined by how many people want to pay for it!
John S.
You can add a Quickie to increase the gain.mcandmar said:
You can add a BeePre to obtain a set of balanced outputs.mcandmar said:
mcandmar
New member
John Swenson said:
That sounds like a whole other project then. I was just curious if the DAC chip had multiple channels that could be used to create two inverted outputs. For example some ES9018 implementations feed multiple channels into an I/V board to create a balanced output, hence i was wondering if the same can be done via an add on board.
Caucasian Blackplate said:
Yeah but no.
Caucasian Blackplate said:
Never knew it could do that, interesting bit of kit.
