Ummm, yes. Imagine watching paint dry.... well it would be quicker than this project!
I have several new designs that are in the "output" stage (making Gerbers and documentation, readme files etc).
uDSP is now on GitHub. It is not tested yet, so why do I put it online? Well the files are "released" and nothing I release has any warranty so it is a convenient place to keep track of released files. If you don't have some method for tracking revisions it can get messy very quickly. I was trying to use BitBucket with private repositories, but running two Git systems is also messy.
Oh and what is uDSP? Well it's a 5cm x 5cm version of the other DSP board I have released. Same hardware, all DSP connections available (minus the thermal diode, although that is on test points).
Why? Not sure, but it seemed like a good idea. Not designed to work with ribbon cables, thinking of more of a stack of boards. Like many other popular hobby boards. I shall order some soon to play with.
Only possible issues, I have used the regulator built into the DSP and cannot work out if being held in reset disables the regulator. The reset circuit works off this regulator, so if the regulator doesn't work in reset then the reset circuit won't work.... Gives me a headache thinking about it.
Oh so KiCad 5 relative paths do still work, but you have to specify the base directory to start, it does not assume the current project directory.
Having updated (and broken and then fixed again) some other projects I am working on I at least know how it should work, and how to break it if I do it in the wrong order...
Just a lot of projects to update, and new documentation needed to support this.
So KiCad 5 came out whilst I was working on new layouts. It's pretty good, some limited regression over version 4, but otherwise a good improvement. I am porting all projects to it.
The new schematic library management is fantastic, and yet they did not do the same to the PCB libraries. The difference being that in the schematic libraries every project library and component is accessible from a panel with a tree structure. The PCB libraries are still driven through dialogues. Consistency is key with these type of programs so a little disappointed here.
However I am seeing some issues taking my design files across multiple machines. I store the designs on my online cloud storage. They should open and work on any machine with relative paths for libraries. This used to work. However this is not working with version 5. It is causing me some issues, and I would like to resolve it before updating, but it does not seem consistent between machines.
Other plans; well I am not happy with the website, too much text and not enough content that is useful. Would like something a little more streamlined.
Once I get around to ordering new PCBs I would like to make more content. How to build them. How to use them. Some guides (video and PDF) on how to use SigmaStudio. Some useful demo applications. Not sure on timescales, have a lot more than this to work on in life!
Just a quick update, have finished various layout changes. Just need to produce manufacturing data, which will likely show a few issues that need tweaking before sending them off.
New designs are:
CoreOne Rev 2.0 - XMOS200 based board with Ethernet, USB, I2S, I2C, SPI. This is simplified over my first attempt, as the last version was very cramped and easy to knock off components you just fitted during build. Also modified the thermal pad so should be easy to solder with a soldering iron, and hopefully retain electrical performance.
An AK4458 DAC - same style as the existing ADAU1966 DAC but only 8 channels.
Also updates to the ADAU1966 DAC board and the ADAU1452 DSP board.
Any new/modified designs are now in KiCad 5.0.0 which is a good upgrade over version 4. I will be updating other designs so they are all in version 5 to keep things consistent.
So, end of summer is approaching. Didn't' achieve anything I planned this summer. Moving home obviously didn't help.
Promised myself I wouldn't rant on here too much, but suffice to say my job sucks. Our competitors products are twice the volume of ours, for good reason, but when you've been getting away with something for a long time.... Too much said already. Hindsight is a wonderful thing, but any of my previous jobs would be much much better in every possible regard. If anyone wants to employ an experienced and competent electronics engineer, with 9 years industrial/commercial experience and 25 odd years hobby experience, I did start when I was 8 or so, then contact me. Pretty easy to find on Linkedin, one of the first UK results to appear. I am happy to give anything a go, and pretty good at stuff I haven't done before as well, can hit the ground running as you might say.
*End of rant*
Back to the topic of OHDSP, I am almost done with some layout and component changes on a lot of boards. A new 8 channel AK4458 DAC design is complete as well. I have reworked my XMOS based board, the last one had everything and the kitchen sink, and all components on the top side. I thought (or hoped) it might be something I could develop and sell in products/as a product. However after destroying two £15 XMOS micro-controllers, and moving, that was shelved for a while. I am not doing this for money so I reverted to a design similar to the existing ones. Less practical for mass production, but that doesn't matter.
The new XMOS design is still cramped, and I have tried numerous layouts on a 10cm x 10cm board, it's not ideal (no PCB ever is!) but it's not terrible. Just deciding on the ground connections to the exposed pad. I will probably buy a hot air station. My ADAU1452 DSP board has solid connections, or I thought it did, to ground and that worked fine. I think the issue is that I used low temperature solder paste, just with a standard soldering iron, to get the centre pad started and I didn't do that with the XMOS chips. But it was several years ago I last built an ADAU1452 board.
Actually on the ADAU1452 (1451, 1450) board the ADAU1466 and ADAU1462 parts are compatible as well, so I have added these part numbers in the new design, although they should work fine on the existing boards, but this is not tested.
I am also working on some other PCB layouts, not related to this. I really want some nice "big" amplifiers. I have a pair of B&W CM10 from previously working there, and even my homebuilt speakers, and they both could do with a bit more than 40W-50W sticking up them. Some of the older Yamaha (MX-800, MX-1000, MX-1) amplifier look interesting, with service manuals online, but eBay purchases always work out badly for me. Also there is the SC480 and Studio 350 from Silicon Chip that look like good, high performance, simple designs. Similar to the old Cyrus One/Two, which I really like, in design. I have been playing with layouts for all of these - for personal use only - so that has been taking up time. You can never have too much power 😊
Not exactly true, but living with many boxes around me. Some of which have OHDSP hardware in. So I have moved home. Next step is to unpack, eventually.
With this project I have been starting the CoreOne layout from scratch. To be honest the pin layout of the XMOS chip is annoying with regards to the Ethernet and USB connections being on opposite sides of the package.
I am sure most devices with USB and Ethernet would want them in the same place on the PCB. Most likely at the outer edge of the housing/enclosure. Not sure why they are at opposite sides of this chip.
I have a number of other updated layouts which have been ready for a while. As this is a hobby project, and I am not rich, these are just waiting for the right time to get them manufactured.
Anyway, to anyone reading this, this project is very much still alive. After a couple of recent emails yes what exists does work, if it didn't I would remove the designs. Remember I am not selling a commercial project, this is a selection of designs which to the best of my knowledge work as expected but be prepared to do some work if you want them.
Or it is supposed to be here in the UK. Maybe it's taking a holiday this year.
I am now moving again, so time to pack everything up. I had been looking at my DAC boards and trying to work out why the common mode voltage seems wrong. Or more specifically why it seems to be close to the tempreature sensor reference (TS_REF) voltage. The pins are labelled as per the datasheet, and the TS_REF output is at 0V which is also not normal with default registers.
I may get a little more time to play before I need to pack this stuff away, but ideally I need to build up a new board with just the DAC chip in place and work from here.
Belated Happy New Year to any one reading this before January runs out.
I have not done much - if anything - on this since November. I had also forgotten where I was.
So I have an updated DACOne design - this is using the 16 channel ADAU1966 or ADAU1966A. I still need to track down the issue with the CM voltage on the current design being lower than expected.
I have a new DACTwo design almost ready. Given the ADAU1966(A) parts are now hard to find or expensive, this new design is an 8 channel design using the AKM AK4458 device.
I am still working on the CoreOne XMOS design. I had a new layout, but the track work wasn't good. The USB and Ethernet are the big issues. I would like to keep them on the same edge for mounting in a case, but the device pin out puts these on opposite sides of the IC. Other track work was long with the connector placement and PSU layout I had used so I am redoing this again.
There are other issues I am seeing as well now with component supplies, both in my hobby and professional design work. Generally electronics components lead times are long now. Lead times for some simple passive parts can be 3 to 6 months or longer in some cases.
It has been having an affect when trying to design for my hobby projects, or even volume quantities. You may choose a suitable, fairly generic part with appropriate size, cost and foot print only to find, by the time you wish to build it, that every supplier has low or no stock and you have to wait 3 to 6 months or change the design.
I have limited development and play time for this project at the moment, and for the foreseeable future. However it will continue. I have another project that I have been working on for nearly a year now. I need to get finished in the next few weeks as it is not for me. Once done that will free up some time and space,
So the current ADAU1966 boards I am playing with have a new issue I have discovered. The ADAU1966 generates it's own "common mode" voltage reference to bias all the outputs to 2.25V. Two of the DAC boards I have tested at 1.47V on the CM pins.
Something is wrong here, it may be the extra capacitance I have on the pin. The data sheet recommends 10uF and 100nF. I have 100nf, 22uF (ceramic) and 220uF aluminium electrolytic. More is always better!!! Or not, in fact that is complete rubbish. I shall remove the 220uF and see what happens.
Extra PCB footprints are not an issue. It's all there to experiment with. If the changes are needed they are not really unexpected as with any project in development.
In any case, and I think I pointed this out in a previous post, the ADAU1966 seems to be hard to get now. The newer ADAU1966A is twice the price and just as hard to find in stock. Even so it is still a good DAC. I am working on a revised PCB layout, it would be good to make sure all designs (new and old) and up to their best performance and working 100%
So after much head scratching the EEPROM selfboot is working again. The SPI address doesn't seem to matter as long as the EEPROM settings are correct. MP0, MP1, MP2, MP3 pins also need setting correctly - although I have changed these a few times before they are correct as it's not completely obvious.
This all needs documenting properly. Hopefully with this box of tricks now working, almost as expected pending a reduction in analogue output levels, I can start doing that.
Other reworked PCB layouts are still work in progress, a few more tweaks needed to finish them off.
An electronics engineer and a long term electronics hobbyist. I like tinkering with stuff and making things.