Something I have been working on

[dioioib]

The good news is the boards are in Alaska, so I should see them by the end of the week or perhaps early next week.

The first 4 prototypes with be finished by the end of the following week, and the code base will be refined further. I would have to say I have nothing more than a basic framework for the code, but the I/O will not be that difficult to finish as SD card read/write are complete, midi I/O is mostly ready, but of course untested. The synthesis engine will be something I am looking forward to developing.

I will post some photos of the boards when they arrive, and after they are populated.

[Inventor]

Way cool dio, looks like you're making excellent progress!

Les

[dioioib]

[Inventor]

dio, do you have any updates on your project? It's been a month since you posted about it.

Les

[dioioib]

Sure do.

Errors.

There were some serious errors on the first run of boards, due to incorrect pad sizes and spacing. The TQFP-100 I used and the TQFP-100 Microchip uses are completely different. So I have taken the month to redesign the board.

The board now has +5/-5 and +3.3v power, I am designing an amplifier section which will be suitable for high-pass or low-pass in the 5th order. And the DAC will be connected correctly this time. The I2C connector was connected to general i/o and not the I2C interface on the PIC. That error I should have caught the first time.

The second version has much larger pads for accessing every additional pin on the pic. With spacing which would allow for a pin header to be soldered to the extra pins.

The prototype area is much larger including two additional areas on the front and back side of the board. And the pads in each area are larger which will make it wasy to hand solder parts onto them.

There have been general optimizations across the entire board, better layout, gound planes for 5v and 3v to separate them. I am still trying to decide if I should plan to use decoupling capacitors on the board but I think the design won't require this.

So thats what I have done so far. The next run of boards should be on within the next few weeks. I have ordered more DAC's from Mouser, as DigiKey is out, and the remaining parts have arived from there suppliers. So things are still moving along although a litter slower, as I am taking more care in designing the second run.

[Inventor]

Good to hear the news, dio. I am also doing a second run of the board that I am developing at the moment. One question I have is: "Why do you think the board doesn't need decoupling capacitors?

Les

[dioioib]

Well good question and the only answer would be that I have been able to get decent results with out them.

I have plenty of bypass caps on the board to smooth power out, but as for decoupling I have never really needed them.

In the past I have used decoupling only for audio inputs and a few supplies I did not want directly connected. Whats your advice Inventor, I know you have spent a great deal of time in the field and your recommendation is appreciated.

[Inventor]

Oh, it's a terminology thing. I've always heard the power supply caps called decoupling caps. I thought you were not using supply bypass caps when you said you didn't need decoupling caps. My mistake I suppose. Sounds like you have the situation well in hand. Good luck!

Les

[electri-fire]

[dioioib]

So a little update. The new boards are finally redesigned. I am looking forward to placing another prototype order this week. I have decided to rework the prototype area again and adjust the pad size making them larger and easier to work with by hand; in addition the area was expanded.

---

pic32_april5-2010.png
Description:
Filesize:	138.23 KB
Viewed:	679 Time(s)
This image has been reduced to fit the page. Click on it to enlarge.

---

[dioioib]

Prototype run #2 just sent today.

I am finally satisfied with the board design for the second run. This one is looking good. I'm looking forward to getting these error (fingers crossed) free boards. I will keep everyone posted.

[Inventor]

Good news, dio! I hope your boards work perfectly! As I mentioned in chat, there has been some interest expressed by someone who is planning to offer a new line of unique modular products. Your board is really cool because hopefully we can set it up so that a customer can download a variety of different program selections to reconfigure your module.

Of course, you've got to get the boards working, write a USB bootstrap loader, and start coding up those programs first so there is still a way to go on your project. My commendations in your persistence on this project, I really think it will sell well.

Les

[dioioib]

Ok I need to confer with the community here. I just got the quote back for the next prototype run. We are looking at an increase in cost by about 2x because of the new specs. I might be able to retune the boards but we are looking at fairly tight specs with all of the stuff crammed on there and it being 4-layers increases the price too.

The pricing of the boards I was shooting for between $50-100 but if I order quantity I am sure the price will not reach anywhere near $100. What do we think?

[Dan Lavin]

Are you saying the protoboards are for $100-200 and in volume much less than $100? If that's so, I don't think that's out of line. But i know a number of folks here might have a little trouble with the proto pricing.

[dioioib]

right now I am talking about prototype price. My target price for a complete unit was intended to be around $100-$150(CAD) populated. Or at least that was the objective. With an increase of double to board cost alone I would likely be out of my target by far too much. I was contacted by the PCB manufacturer and I might be able to reduce a majority of the cost by increasing via size and changing trace width. The tight specifications are driving up the cost. So I have decided to change them, I was shocked by the price to be honest, but after finding out what the issues were I realized hey we can easily change that. There are only 201 vias on the new board and most of the parts are SMD so there is some room to work. The area of real issue is the PIC32 area where the pad spacing is very tight to begin with, after adding bypass caps and vias to he underside of the board we are left with little remaining space. This will delay the prototype run but reduce cost. From the beginning my objectives were affordability, modularity and flexibility. I do not want to lose site of that.

Thank you for the input.

[Inventor]

I feel that cost is a critical factor. Anything you can do to reduce cost will help greatly. Is there other packaging for the PIC32? Can we move all parts to the top of the board? I don't really see the need for the prototyping area really, as few people will make use of it - would removing or reducing the prototype area(s) help lower cost?

Les

[dioioib]

Inventor, I wish there was another package for the TQFP100 with a larger pin spacing but as it turns out there is not. That was the major error on the first run of boards, I had to rebuild the board in a rush and used a standard TQFP100 pad spacing of 2mm and the PIC32 uses 1mm. Lesson learned.

The adjusted files are at the board house, and I am awaiting a response.

I understand the idea of removing the prototype area. However, I am going to keep it; the board will benefit from the area and stoke creativity and tinkering with the hardware. As the board is designed right now the prototype area is only filling void space, and this is really not the major cost increase on the board. Additionally, I have considered the idea of using the space to fill out the I2C bus, we can connect up to 128 devices if my memory serves me, and really the board only uses about 3. So if required the area can be used for that type of expansion.

The board does have a majority of the components on the top side with the exception of a few capacitors. The DSP on the underside is not really ready yet but I can not justify moving that to the upper side because it will increase the board beyond the desirable size.

[Inventor]

OK dio, sounds like you've thought things out well and I know you're on top of your game! I Hope all goes well with this run of boards...

Les

[dioioib]

This run I have taken more care. There might be something I am overlooking but I have done everything from a dry run for placement, double-triple-quadruple check of the pin connections, although I might check them one more time. So I am feeling good. Your input is always appreciated though, I enjoy questioning the design and why I have done things as such. There is always room for improvement.

[rubendelacosta]

looking good!
can't wait to see this board working/playing

[dioioib]

Ok so after retooling the board the cost has dropped on the prototype so it is in a more reasonable price point for the run. I'm going to review the board one more time and see if there are any other cost saving tweaks I can do. But the setup fee has dropped about $100-200 so I do not think there is much more room to go. Worth a shot though.

[dioioib]

Order has been placed so now we wait...

In the mean time I have two other projects I need to direct attention two. But I should have the boards in about two weeks. I will keep everyone posted.

[dioioib]

Good news is the new boards have arrived. I will be populating the boards over the next few days when time permits. They look great and I have a feeling they are going to work great. Problem is I have to solder SMT by hand which can take a little time and finesse. However, as it stands we are looking at getting the code base working over the next few months. After which time I will work on control surfaces for the device and other doodads.

[Inventor]

That's good to hear dio, hope your boards work flawlessly!

Les

[Taylor]

This is really interesting. I'm really looking forward to hearing how it's coming.