Title: Hardware manufacturing update
Post by: ian on February 06, 2010, 09:38:39 am
Post by: ian on February 06, 2010, 09:38:39 am
Just stuffed the first PCB for a new minor manufacturing revision of the Bus Pirate. No functional changes, just some tweaks based on the past few months of production. This PCB will probably be used after the current batch is depleted:
1. Replaced 3x 2.2uF tantalum capacitors with 3x 1uF ceramic. This is moderately cheaper, uses smaller parts, and it uses less tantalum (a conflict mineral).
2. Connected PWR and USB LEDs to 3.3volt PIC supply instead of 5v usb supply, now all LEDs are supplied by 3v3. This is an additional test of the power supply, the LED brightness is more consistent, all LEDs now point the same way, and all LEDs use the same resistor value.
3. LED brightness has been an issue. Future revisions will have 1K1 resistors instead of 390R, the LEDs are much dimmer.
Seeed is making a large batch of v3a Bus Pirates that will hopefully break the endless preorder cycles. Except for a brief period in December, the Bus Pirate has never been available for immediate delivery. It's always sold out on back order before the solder has even reflowed. That will change at the end of the month.
1. Replaced 3x 2.2uF tantalum capacitors with 3x 1uF ceramic. This is moderately cheaper, uses smaller parts, and it uses less tantalum (a conflict mineral).
2. Connected PWR and USB LEDs to 3.3volt PIC supply instead of 5v usb supply, now all LEDs are supplied by 3v3. This is an additional test of the power supply, the LED brightness is more consistent, all LEDs now point the same way, and all LEDs use the same resistor value.
3. LED brightness has been an issue. Future revisions will have 1K1 resistors instead of 390R, the LEDs are much dimmer.
Seeed is making a large batch of v3a Bus Pirates that will hopefully break the endless preorder cycles. Except for a brief period in December, the Bus Pirate has never been available for immediate delivery. It's always sold out on back order before the solder has even reflowed. That will change at the end of the month.
Title: Re: Hardware manufacturing update
Post by: Sjaak on February 06, 2010, 11:28:14 am
Post by: Sjaak on February 06, 2010, 11:28:14 am
Nice! good to hear the backorders will be over.
You did switch the colors of the vreg and mode. Did you do it on purpose? I would suggest the mode and vreg are supposed to be red (danger) I was surprised that the mode led on v3 was green (after fixing it) (no big deal though)
Is the footprint the same as v3? (to make stackable busconvertors possible) Also will the isp connector be upright in the final design?
You did switch the colors of the vreg and mode. Did you do it on purpose? I would suggest the mode and vreg are supposed to be red (danger) I was surprised that the mode led on v3 was green (after fixing it) (no big deal though)
Is the footprint the same as v3? (to make stackable busconvertors possible) Also will the isp connector be upright in the final design?
Title: Re: Hardware manufacturing update
Post by: ian on February 06, 2010, 02:50:48 pm
Post by: ian on February 06, 2010, 02:50:48 pm
The colors were random, I just picked alternating colors from my parts box.
The next revision will come without an ICSP header because the new jump-to-bootloader command is available in the firmware.
Yes, it is the same footprint. The number one priority was to leave the board dimensions exactly the same. I'm redesigning the HVP adapter board to be stackable, at your request :)
The next revision will come without an ICSP header because the new jump-to-bootloader command is available in the firmware.
Yes, it is the same footprint. The number one priority was to leave the board dimensions exactly the same. I'm redesigning the HVP adapter board to be stackable, at your request :)
Title: Re: Hardware manufacturing update
Post by: Sjaak on February 06, 2010, 04:16:15 pm
Post by: Sjaak on February 06, 2010, 04:16:15 pm
Great! now i want definiately one!
Just for the bigger audience, a stackable pcb won't fit into the case seed is selling, unless they use a longer header (seed does sell a nice long one). I prefer a header on the isp for starting the bootloader, I do have such a nice red jumper :P (it is also more stable for stacking)
Just for the bigger audience, a stackable pcb won't fit into the case seed is selling, unless they use a longer header (seed does sell a nice long one). I prefer a header on the isp for starting the bootloader, I do have such a nice red jumper :P (it is also more stable for stacking)
Title: Re: Hardware manufacturing update
Post by: Scorpia on February 07, 2010, 11:42:21 am
Post by: Scorpia on February 07, 2010, 11:42:21 am
Ian,
nice to see some changes and the more stable manufacturing coming along.
I must admit i have been wondering about the capacitors on your boards lately . My feeling is i am wondering if there is enough capacitance for all occasions.
now dont take this to serious as its only me wondering. i havnt looked into this in a technical way. just a gut feeling after a design i had needed upgrades in the capacitor sizes for stable operation.
nice to see some changes and the more stable manufacturing coming along.
I must admit i have been wondering about the capacitors on your boards lately . My feeling is i am wondering if there is enough capacitance for all occasions.
now dont take this to serious as its only me wondering. i havnt looked into this in a technical way. just a gut feeling after a design i had needed upgrades in the capacitor sizes for stable operation.
Title: Re: Hardware manufacturing update
Post by: ian on February 07, 2010, 01:56:26 pm
Post by: ian on February 07, 2010, 01:56:26 pm
I welcome any advise. The Bus Pirate has the 10uF supply source cap recommended by both the MIC5202 vreg datasheet and the 232R datasheet. The first Bus Pirates had 10uF output caps on the MIC5202's, v3a has 2.2uF, but this was because of what I had in my parts box and not because of the datasheet. The datasheet recommends >=1.0uF, so I thought I'd use the smaller, cheaper parts.
There's certainly room for improvement, and my values based on datasheets won't ever perform as well as considered and calculated values based on desired tolerances.
If you have any suggestions, please let me know.
There's certainly room for improvement, and my values based on datasheets won't ever perform as well as considered and calculated values based on desired tolerances.
If you have any suggestions, please let me know.
Title: Re: Hardware manufacturing update
Post by: pppd on February 07, 2010, 09:58:27 pm
Post by: pppd on February 07, 2010, 09:58:27 pm
Will the new BP PCB be green or blue? I was thinking of getting a second BP for RF modules testing, but I would like a different colour than red :-)
Title: Re: Hardware manufacturing update
Post by: Sjaak on February 07, 2010, 10:01:05 pm
Post by: Sjaak on February 07, 2010, 10:01:05 pm
pirateblack! :P
Title: Re: Hardware manufacturing update
Post by: artg on February 08, 2010, 11:34:23 pm
Post by: artg on February 08, 2010, 11:34:23 pm
In order to fit the BP (I have a v3a) more neatly in a case, I've removed the ICSP header and replaced the I/O header with a rightangle connector. I can always put the ICSP back if I need it, but I see from the comments above that you're removing it anyway.
I notice that you've use the large ftdi232r package. This has a number of unused pins (cbus and part of dbus) that can be used in bitbang mode. If some of these were connected to the ICSP pins, would it be possible to program the BP bootstrap in without needing external ICSP at all ?
I notice that you've use the large ftdi232r package. This has a number of unused pins (cbus and part of dbus) that can be used in bitbang mode. If some of these were connected to the ICSP pins, would it be possible to program the BP bootstrap in without needing external ICSP at all ?
Title: Re: Hardware manufacturing update
Post by: ian on February 09, 2010, 08:04:40 am
Post by: ian on February 09, 2010, 08:04:40 am
Great idea to bootstrap the PIC from the FTDI chip.
Is there a smaller 232R chip? I thought there were only 28SSOP and 44QFN.
Is there a smaller 232R chip? I thought there were only 28SSOP and 44QFN.
Title: Re: Hardware manufacturing update
Post by: artg on February 09, 2010, 11:14:37 pm
Post by: artg on February 09, 2010, 11:14:37 pm
There's a 32QFN package, about half the size of the SSOP.
But actually, I was thinking there was another FTDI serial chip without so many GPIO pins. In fact, the others all seem to have more.
But actually, I was thinking there was another FTDI serial chip without so many GPIO pins. In fact, the others all seem to have more.
Title: Re: Hardware manufacturing update
Post by: bikenomad on June 04, 2010, 05:24:24 pm
Post by: bikenomad on June 04, 2010, 05:24:24 pm
[quote author="ian"]
1. Replaced 3x 2.2uF tantalum capacitors with 3x 1uF ceramic. This is moderately cheaper, uses smaller parts, and it uses less tantalum (a conflict mineral).
[/quote]

I think that ceramic output caps may be a bad choice for the MIC5205 family because of their low ESR.
According to the MIC5205 datasheet, tantalum output caps are required.
They say on the first page:
And, about output caps:
1. Replaced 3x 2.2uF tantalum capacitors with 3x 1uF ceramic. This is moderately cheaper, uses smaller parts, and it uses less tantalum (a conflict mineral).
[/quote]

I think that ceramic output caps may be a bad choice for the MIC5205 family because of their low ESR.
According to the MIC5205 datasheet, tantalum output caps are required.
They say on the first page:
Quote
For low-dropout regulators that are stable with ceramic output capacitors, see the μCap MIC5245/6/7 family.
And, about output caps:
Quote
The output capacitor should have an ESR (effective series resistance) of about 5Ω or less and a resonant frequency above 1MHz. Ultra-low-ESR capacitors can cause a low amplitude oscillation on the output and/or underdamped transient response.