Anyone have a north american sourced parts list for the HCFR colorimeter sensor?

check the main HCFR free software thread. Some did. Also check after those posts for one of mine. I found a source in partsexpress.com for the abs case.

The OP of that info was StephenR0, that'll help limit the pages you ahve to go through.

jeff

greeno,

Found it, thanks!

Is there a source out there that can provide the PIC controller already flashed?

yeah, that would be great, I am looking for help with flashing as well...

In that thread, someone went through the same issues. I don't recall what they finally did. That's what's keeping me from trying to build it...

Best,
jef

Which PIC are they using, I might start flashing them.

U1 - PEAK 18F2550 (Microchip) (ref. Farnell 9321250)
1 x $7.83@ Mouser 579-PIC18F2550-I/SP PICmicro - PIC18Fxxx Flash MCU's 32kBF 2048RM FSUSB2

There's a discussion somewhere in the hcfr software thread about that after page 22.

What would a printed circuit board (ready to solder) and a programmed PIC be worth to the DIY'ers?

Are you thinking about putting together a "kit" that would include the programmed PIC and the PCB? I think float a price and see how much interest there is...

It would be worth at least 20-25 USD to me. Wonder what would be the postage to Norway??? and how can I pay you?

Are you thinking about putting together a "kit" that would include the programmed PIC and the PCB? I think float a price and see how much interest there is...Yeah, thought about it. Quality PCB (but w/o soldermask or silkscreen) would cost about $15e. I think that particular PIC would be about $7-8. Call it $35 after shipping to me. Shipping out +$, Labor +$, full kit parts ++$. Looks like not enough margin to fool with it based on one person chiming in with an offer of $10 under cost :).

I'm interested.

I meant chip only, have no clue of what PCB costs. If we talk about full kit, that is another story. I have an opportunity to find other parts locally, chip programming is what keeps me on hold... But if you come up with complete kit, I will most certainly consider that.

I would be interested in a programed chip and PCB.
Maybe a complete kit would be OK.

What would a printed circuit board (ready to solder) and a programmed PIC be worth to the DIY'ers?

@GetGrey, are you going to come up with price offer?

Maybe later. Have hands full right now with www.cineslide.com and www.calibrate.tv.

Programming the PIC is no biggie, but to make it worth the trouble, I need to mate it with the PCB IMO. Everything else anyone could get easy enough.

If someone wants to do the thread reading homework and PM or email me the direct links to:
- the PIC code (hex, asm, whatever)
- the PCB layout
- the parts list

I'll see if I can squeeze it in. It's not that I'm too lazy to do it myself, I'm just in the weeds. I have the skills and tools to do it for folks though, easy for me once setup so though I might captialize on it.

I don't kow if there are any shipping restrictions for PICs overseas (from USA), have to look into that. Anyone know chime in with link to answer?

Here's the link to the HCFR page: http://www.homecinema-fr.com/colorimetre/sonde.php

Go down to the pdf for the PCB (it's downloadable).

The hex code for the PIC is availble at the very bottom of the linked page - click on "ici", that's a hyperlink, in the sentence that begins: "Le PIC 18F2550 devra être..."

Here's the parts list taken from StephenR0's post in the original HCFR sofware thread:
C1, C3, C4 - 100nF (CMS 1206 or trad. 5.08)
3 x $.09@ Mouser 581-SR215E104MAR Radial Monolithic Capacitors 50V .1uF 20% Lead Free

C7 - 220nF (CMS 1206 or trad. 5.08)
1 x $.26@ Mouser 581-SR205E224MAR Radial Monolithic Capacitors 50V .22uF Z5U 20% Lead Free

C5, C6 - 33pF (CMS 1206 or trad. 5.08)
2 x $.06@ Mouser 140-50N5-330J-RC Ceramic Disc Capacitors 50V 33pF NPO 5% Tol

C2 - 100µF 10v (Chemical Polarized)
1 x $.06@ Mouser 140-XRL10V100-RC Radial Electrolytic Capacitors 10V 100uF 20%

R1 - 3.3K
10 x $.04@ Mouser 291-3.3K-RC 1/4W 5% Carbon Film Resistors 3.3Kohms 0.05

R2 - 150R
10 x $.04@ Mouser 291-150-RC 1/4W 5% Carbon Film Resistors 150ohms 0.05

R3 - 2.2K
10 x $.04@ Mouser 291-2.2K-RC 1/4W 5% Carbon Film Resistors 2.2Kohms 0.05

R4, R5, R6, R7 - 0R (Straps)
Cut off resistor leads.

R8, R9 - 27R
10 x $.04@ Mouser 291-27-RC 1/4W 5% Carbon Film Resistors 27ohms 0.05

D1 - Led Green 3mm or CMS 1206 (rise with dimensions copper)
1 x $.13@ Mouser 604-WP132XPGD LED Standard GREEN DIFFUSED

D2 - Led Infra-red 5mm (gone up crossing)
2 x $.21@ Mouser 782-TSAL6400 Infrared Emitting Diodes 25 Degree 25mW

SW1, SW2 - monostable Switch 5.08 * 7.62 (Microphone switch)
2 x $.26@ Mouser 688-SKHHAJ Tactile Switches 6.0x6.0x4.3mm 100gf

X1 - Quartz 4Mhz
1 x $.36@ Mouser 815-AB-4-B2 HC49 MHz Range Metal Can Crystals +/-20ppm 4MHZ

Q1 - NPN - BC547
1 x $.06@ Mouser 512-BC547BTAR Small Signal Transistors NPN Si Transistor Epitaxial

P1 - Standard Base plate Ci USB B (ref. Farnell 152432)
1 x $.71@ Mouser 806-KUSBX-BS1N-B USB Connectors B TYPE RECEPTALCE BLACK

U1 - PIC 18F2550 (Microchip) (ref. Farnell 9321250)
1 x $7.83@ Mouser 579-PIC18F2550-I/SP PICmicro - PIC18Fxxx Flash MCU's 32kBF 2048RM FSUSB2

U2, U3 - Sensor TAOS - TCS230 (CMS SO8, ref. Farnell 4891211)
2 x $5.69@ Mouser 856-TCS230D Color Sensor TriColor Sensor LTF

K1 - Connector Mono Jack 3.5 for Ci (ref. Farnell 152206)
1 x $.70@ Mouser 161-3412-EX Phone Jacks PHONE 3.5mm MONO

1 Support 28 Pins (ref. Farnell 4242403)
1 x $1.22@ Mouser 649-DIP-328-001B DIP Sockets 28C OPEN FRAME

1 Case ABS, 65*100*20 mm
[Hammond clipper series: 105mmx60mmx22mm part number 001104, $8.74 <<< this is probably too expensive???]

1 USB cable A-B 4 Meter
1 x $5.52@ Mouser 571-1487598-1 USB Cable Assemblies USB, A-B, 25/20 BLACK 4 M

1 Mono audio cable 120 inches
1 x $2.78@ Mouser 172-1281 Audio Cords 3.5 MONO PL-S/T 120

I'd be interested in 2 maybe 3 programmed PIC chips for starters.

May be 1 or 2 PCBs too. Otherwise I'd just end up sourcing it out to one of those online companies.

Hell, I'd take a mark-up on the rest of the components for the convenience.

I have an offer for someone who is willing.
I just found my pic programmer from a project a few years ago....now I would be willing to program a pic for feedback on if it worked.....I have not used the programmer for about 3 years, so not sure if it is functional......you must supply pic as I don't have any.
PM or email me if you are willing to be a guinee pig :)
If it works.....I will be offering pic programming.

Later
RayJr

I’ve been impressed with HCFR (Thanks Guys!) and have been interested in trying out their probe based on the TAOS YRGB photo-diode sensors. I finally got around to finishing a few of these. I decided to do a SMT version of the board as it’s actually easier for me to program PICs in-circuit. While I was at it, I also implemented another design that I had that uses a smaller PIC on the same board. (That’s what all the extra non-populated footprints are about.)

Other than being all SMT, the other changes I made was to move the photo diode arrays closer together and rotate one 90 deg. The mounting hole spacing is for a 2.6” square plastic enclosure however DigiKey screwed up and sent the wrong ones. I’ve also got some IR, UV filters / dichroics and some dispersion films I’m looking forward to playing with.

My initial impression of their performance is pretty positive. It does indeed seem to be much better down low and doesn’t go ‘color blind’ or bonkers <20 IRE like my spider2 sample does. I’m also surprised about the unit-to-unit variation. I’ve tested five of them and they all perform identically as near as I can tell.

Anywho, I’ve got a few extra assembled, programmed (HCFR 5.1), and tested units (with boxes?) and / or a few raw boards (if you’re comfortable with SMT (0603) assembly.) if anyone is interested. (US) It’s a bit of an effort and expense to go through for just one. :D

I can also put together a Bill of Materials (North American suppliers) and gerber files if there are any other masochists. Send me a PM if you’re interested.

Dave

Great work Dave ! ;)

And I think you well spotted the pros/cons of it... Main cons being the quite wide angle (unless you use something to direct light), a bit of IR sensitivity (need an IR filter), and, obviously the need of calibration files (at least for now :))...
On the pros side you have the very low cost and fast response time (although many measure are taken and averaged) as well as quite good result in low end.

It's really nice to see an alternative design ;)

--Patrice

What pic does this new design have? I am still struggling to find someone to program 18F2550 PIC....

What pic does this new design have? I am still struggling to find someone to program 18F2550 PIC....
It uses the same PIC (and 5.1 hex file). It is surface mount (SO28) rather than through hole (DIP28) and I program it in-circuit via J2.

Dave

Attached are the gerbers and Bill of Materials for the HCFR style probe board I did for anyone who wants to roll their own. The necessary hex file and USB driver can be found on HCFR DIY page. May the force be with you.

One other deviation from the HCFR designd that I forgot to mention- I didn’t include any of the parts for the IR emitter (used to talk to a set-top DVD player when using the HCFR pal DVD) but I did include footprints and connectivity for it. The board should support it if you chose to add it yourself. I haven’t tested that part and don’t plan to.

Dave
QNTY Reference Description DigiKey Part # Vendor Vendor Part # Ea $ Tot $

6 C1,C2,C4,C5,-C7 CAPNP,0.1uF,50V,0603 490-1519-1-ND Murata GRM188R71H104KA93D 0.07 0.42
1 C12 CAPP,47uF,10V,TANT 399-3757-1-ND Kemet T491C476K010AT 0.52 0.52
1 D1 LED,GREEN,SMT0603 160-1183-1-ND Liteon LTST-C190GKT 0.09 0.09
1 R1 RES,10K,100mW,1%,0603 P10.0KHCT-ND Panasonic ERJ-3EKF1002V 0.09 0.09
1 R2 RES,200,100mW,1%,0603 P200HCT-ND Panasonic ERJ-3EKF2000V 0.09 0.09
1 R3 (BACK SIDE) RES,10,100mW,1%,0603 P10.0HCT-ND Panasonic ERJ-3EKF10R0V 0.09 0.09
2 SW1,SW2 SWITCH,SPST,MOM,PB 401-1786-1-ND ITT KSC421G 70SH LFS 0.51 1.02
2 U1,U3(BACK SIDE) IC,PD SENSOR,TCS230 856-TCS230D(Mouser) Taos TCS230 5.69 11.38
1 U2 IC,PIC,18F2550 PIC18F2550-I/SO-ND MicroChip PIC18F2550-I/SO 10.03 10.03
1 Y1 XTAL,4.00MHz,SMT 490-1200-1-ND Murata CSTCR4M00G53-R0 0.54 0.54
1 J3 CONN,4P,USB-B,RA 609-1039-ND FCI 61729-0010BLF 1.04 1.04
1 PLASTIC BOX HM355-ND Hammond 1593KBK 2.16 2.16
1 PCB 15.00 15.00
TOTAL $42.47

And I think you well spotted the pros/cons of it... Main cons being the quite wide angle (unless you use something to direct light), a bit of IR sensitivity (need an IR filter), and, obviously the need of calibration files (at least for now )...
On the pros side you have the very low cost and fast response time (although many measure are taken and averaged) as well as quite good result in low end. Yea, the need for an initial calibration reference for someone without a display listed in the HCFR calibration files is a con. I was initially somewhat skeptical on how effective the pre-built HCFR probe calibration file could be too. It seems it would rely on very little unit-to-unit variation. After building and testing 5 of them, I’m amazed that I don’t see any unit-to-unit variation. It acts like its totally digital. ;) It may not make a good choice as someones first or only colormeter because of this though.

I’m not sure I’m seeing the wide viewing angle as a con for me now. As you say, it seems easy enough to narrow the view. (A short tube or straw?) I’m also not seeing IR as an issue either (with CRT and LCD) I see the TAOS app note discusses IR filters and I have some filters and dichroics but I don’t see it behaving any differently than my SpyderII in that regard. I do plan on playing more with the optics / filter / diffuser side of it (that’s half the fun?) but it also seems pretty good with the just the raw sensors. It seems most of HCFRs calibration files would rely on a raw (or neutral diffuser) sensor? What displays have you seen IR to be an issue with the raw TAOS sensors?

I’ll be interested to learning more about possibly eliminating the need for calibration files and what IR filters you’re using.


On the pros side you have the very low cost and fast response time
Yea, it is speedy. Little reason to only take 10 points. Low cost? My first one wasn’t so cheap… :D

Dave

...It seems it would rely on very little unit-to-unit variation. After building and testing 5 of them, I’m amazed that I don’t see any unit-to-unit variation. It acts like its totally digital. ;) It may not make a good choice as someones first or only colormeter because of this though.
Yes, that was the start base of our decision to pursue with this design, it is not ideal, but the very (very) low variation from onr to another allow us to do these calibration files and that does the trick... It could be a good first probe if your projector is in the list we have profiles for...

I’m not sure I’m seeing the wide viewing angle as a con for me now. As you say, it seems easy enough to narrow the view. (A short tube or straw?) I’m also not seeing IR as an issue either (with CRT and LCD) I see the TAOS app note discusses IR filters and I have some filters and dichroics but I don’t see it behaving any differently than my SpyderII in that regard.
The wide angle is not a problem when using projectors, but it is clearly annoying on some screen, especially LCDs... But that isn't the kind of device we had in mind when designing this. (we are mainly on the projectors side).

I do plan on playing more with the optics / filter / diffuser side of it (that’s half the fun?) but it also seems pretty good with the just the raw sensors. It seems most of HCFRs calibration files would rely on a raw (or neutral diffuser) sensor? What displays have you seen IR to be an issue with the raw TAOS sensors?
Some projector bulb did emits a few and that could be play a role (Epson TW600/700 are an example) but again it is not generaly a big deal. (and a simple IR filter can do the trick easily... We have also tried a lot of filters, etc... In the end, only a IR filter can help (in some cases) hence the reason we suggest to do the probe as is, using raw sensors...
BTW, IR sensitivity also helped us, as it is the way we use the probe to actually learn IR codes from DVD player remotes and so then ColorHCFR can control player via the probe (embeded IR emiter) and so easier the calibration process ;)

I’ll be interested to learning more about possibly eliminating the need for calibration files and what IR filters you’re using.
To remove the need for calibration, the best way is to get the full real caracteritics of the sensors... That'w what we are working on... Only lack of proper/efficient tools did not allow us to complete this so far.

Yea, it is speedy. Little reason to only take 10 points. Low cost? My first one wasn’t so cheap… :D
Well, 10 points is generaly enough for most people to have a properly setup system, then they can easily go further with ColorHCFR, it allows 20 or any number very easily ;)
I agree first probe is always a bit expensive (trust me, I learned that the hard way with dozens of prototypes :D)

Thanks for your inputs ;)

--Patrice

So I went ahead with ahead to solder myself an HCFR unit. I call this project "Can I still solder", since this is my second attempt in my life to solder with 20 years break in between. I guess I explain all this to make my questions not to sound very stupid :confused:

I started with sensors and man, after just two hours one was in a place.... It really took out sweat out of me to solder those little legs and not glue them together... One day later, I am proudly viewing my creation from all angles and see.... sensor is slightly lifted on oone side. It is about 10 degrees lifting I would say. In any case, it is not perfectly horisontal. Will that effect measuring??? I don't dear to start resoldering process if it is absolutely not necessary. So, does the angle have to be absolutely precise?

Capacitors. Haven't placed them yet. I thought that it is important not to mix + and - when one solders capacitors. But I can't figure out what is plus and what is minus on these small capacitors. How should I proceed?

Does +/- matter for quartz? If so, how should I define what position to put it in?

You won't have to sold them again, angle is not a problem ;) Btw, a easy way of soldering them is to use a small amount of glue to perfectly fit them and then solder.

Quartz don't have polarity :D

--Patrice

Great news :) Thanks for reply; Yeah, glue would have helped, but didn't think about it...

What about capacitors, do polarity matter there?

What about capacitors, do polarity matter there?

Assuming you are building from the original HCFR Bill-of-Materials, C2 is the only polarized capacitor in the BOM.

Dave

Thanks for help dlarsen, yes I follow original BOM. I have put together colorimeter and am almost proud of myself. If it works, will be great bonus :) Is there any way to test it before I plug into my PC? I mean to measure some resistance between USB pins or something similar. I could expect that device will not work due to some mistake in the process, but to blow PC at the same time would be too much... Would appreciate some tips, before I plug it into PC.

If you have an ohmmeter, I’d suggest ohming it out between power and ground at a minimum to make sure you don’t have any power to ground shorts. Also, for first come alive power up, you may consider connecting and powering through a USB hub if you have one. Most USB hubs I’ve torn apart do offer some protection in the form of self-resettable (thermal) polyfuses for the USB 5V. Worst case if something is amiss would be that you damage a $20.00 USB hub rather than you computer.

On first power up, the power LED should chirp fairly rapidly. (look closely, it blinks fast)

Once the driver is installed and the PIC is initialized via communications with the driver, the power LED should glow steadily and only turn off while a measurement is in progress.

Dave

wow, it worked :) I built it! Feels really great. Now is my next step, to learn to use it. Thanks a lot for help.

wow, it worked :) I built it! Feels really great. Now is my next step, to learn to use it. Thanks a lot for help.

Since kits/boards aren't currently available, I'd like to try building a probe using this assembled and tested PIC module (http://www.sparkfun.com/commerce/product_info.php?products_id=762). It has everything except the color sensor, and comes with a bootloader pre-programmed so that programming can be done over USB from any PC without any special hardware.

There are 2 hardware differences to workaround: 20Mhz vs 4Mhz, and pin 13 usage. I'd modify the PIC assembly code myself if it was available, but I haven't gotten a response yet about that. Any ideas/comments?

--Dan

Looks doable to me but you'd need the source as that uses a different target device as well as the pin-out issue.

I asked about PIC source access several months ago via AVS PM, I was advised that they'd (HCFR devs) would look into it and get back with me but I didn't press it.

As to boards, etc, you may check the site listed under my profile...

Dave

Hi
In fact, we prefer not distributing our source code for the probe.
By the way this module seems to be really interesting to get a low cost and easy to program probe (but out of stock)
I will consider modifying the code to fit with it. As we use a similar PIC, it should not be really difficult.
(but at the moment, I have a train in a couple of hours to go snowboarding :D)

Hi
In fact, we prefer not distributing our source code for the probe.
By the way this module seems to be really interesting to get a low cost and easy to program probe (but out of stock)
I will consider modifying the code to fit with it. As we use a similar PIC, it should not be really difficult.
(but at the moment, I have a train in a couple of hours to go snowboarding :D)Thanks Benoit for looking into this. They also have a kit version that uses the PIC18F2550 (http://www.sparkfun.com/commerce/product_info.php?products_id=8265), but it's only $5 cheaper. Your team may even want to contact Sparkfun about adding the color sensor to the board, as they are very open to distributing project boards like this.

--Dan

Thanks Benoit for looking into this. They also have a kit version that uses the PIC18F2550 (http://www.sparkfun.com/commerce/product_info.php?products_id=8265), but it's only $5 cheaper. Your team may even want to contact Sparkfun about adding the color sensor to the board, as they are very open to distributing project boards like this.

--Dan

The PIC 18F2455 and the 20 MHz quartz won't be a big deal. I think it's more interesting to use the mounted version ... the problem is that it is currently out of stock.

Hello everyone,

I too, along with four other friends here in Greece, am trying to build the HCFR colorimeter. We’ve built the PCBs and managed to put together just about everything, but we’re stuck at programming the PIC18F2550-I/SP.

We have the exact recommended Farnell code (9321250) and one of us, who’s rather experienced in electronics, has a programmer which he has used in the past to program a very similar 18-series PIC.
As he told me, his problem is that while most of the hex data (5.1) is transferred normally, the fuses aren’t stored into to the chip.

Are there any hints or special instructions? Any help would be much appreciated.

Hello everyone,

I too, along with four other friends here in Greece, am trying to build the HCFR colorimeter. We’ve built the PCBs and managed to put together just about everything, but we’re stuck at programming the PIC18F2550-I/SP.

We have the exact recommended Farnell code (9321250) and one of us, who’s rather experienced in electronics, has a programmer which he has used in the past to program a very similar 18-series PIC.
As he told me, his problem is that while most of the hex data (5.1) is transferred normally, the fuses aren’t stored into to the chip.

Are there any hints or special instructions? Any help would be much appreciated.
What software is he using with the programmer? I use the official programmers and software from Microchip, and the fuse settings are embedded in the file. I had no problem making 2 colorimeters and programming them correctly.

Edit: OK, I went back and loaded the program in the Microchip MPLAB software. The fuses are shown as follows:

-Full-speed USB clock source selection= Clock src from 96MHz PLL/2
-CPU system clock postscaler= [OSC1/OSC2 Src: /1] [96MHz PLL Src: /2]
-96MHz PLL Prescaler= No Divide (4MHz input)
-Oscillator= HS: HS+PLL, USB-HS
-Fail-safe clock monitor enable= Disabled
-Internal External Switch Over Mode= Disabled
-USB voltage regulator= Enabled
-Power Up Timer= Disabled
-Brown out detect= Disabled in hardware, SBOREN disabled
-Brown out voltage = 4.2V
-Watchdog timer= Disabled-controlled by SWDTEN bit
-Watchdog Postscaler= 1:32768
-CCP2 Mux= RC1
-PORTB A/D enable= PORTB<4:0> configured as digital I/O on RESET
-Low power timer1 osc enable= MCLR enabled, RE3 disabled
-Stack overflow reset= disabled
-Low voltage program= Disabled
-Code protect 00800-01FFF= disabled
-" " " " 02000-03FFF=disabled
-" " " " 04000-05FFF=disabled
-" " " " 06000-07FFF=disabled
-Data EE read protect= disabled
-Code protect boot= disabled
-Table write protect 00800-01FFF= disabled
-" " " " 02000-03FFF=disabled
-" " " " 04000-05FFF=disabled
-" " " " 06000-07FFF=disabled
-Data EE write protect= disabled
-Table write protect= enabled
-Table read protect 00800-01FFF= disabled
-" " " " 02000-03FFF=disabled
-" " " " 04000-05FFF=disabled
-" " " " 06000-07FFF=disabled
-Table read protect boot= disabled

Hello again Thano and thank you for your reply.

My friend who's doing the programming has sent me some additional information.
He says that the problem lies at the addresses 300006 and 30000Β of the configuration bits, as shown below:

http://www.avsforum.com/avs-vb/attachment.php?attachmentid=104936&stc=1&d=1205785998

PIC is programmed on a separate Board (not in circuit programming). He’s using software PicPgm (V.1.0.1 3), and his programmer is PICPgm LVISP (Low-Voltgage, In-System Programmer http://members.aon.at/electronics/pic/picpgm/ )

...and his programmer is PICPgm LVISP (Low-Voltgage, In-System Programmer...

The PIC18F2550 is NOT a low-voltage device. You need the PIC18LF2550 for low voltage (3.3) use. You can flip a 1 to a 0 with a LV programmer but flipping a 0 to a 1 can be problematic. If you can add 5V externally to the programmer that may help but if the programmer is only intended to program LV parts then you need LV parts.

Dave

Hi
Low Voltage programming is not linked to low voltage device. It can be used on PIC 18F2550.

Your problem is that we do not have allowed this option because when you do so, you can not use RB5 port of the PIC any longer.
The difference on 300006 configuration bit is the Low Voltage Programming activation bit.
Your programmer may not allow to deactivate this bit and that can be a problem for the probe.
Can you please make the following test ;
Put your PIC on the probe board, connect it to a PC, press "Reset" and "Load" push together, release "Reset" and after 1 seconde Release "Load".
Does the LED blink ?
Is a new hardware detected by your PC ?
What it it's name ?

Hello Ki,

as my friend told me over the phone, he did as you suggested and managed to install to device as "HCFR colorimeter" or something like that, but it took a few attempts and plugging-unplugging the USB port. He also told me that it seemed to have a somewhat peculiar behavior (appearing and disappearing at the Device manager etc).
I'll have one of the four devices in my hands by tomorrow afternoon, so I can do some further testing myself. Are there any other suggestions?

I think we're making some progress!

If you just plug in the usb connector, it should be recognised as "HCFR colorimeter", but I think the fact that fuses are not correctly programmed could lead to strange behavior.
Then if you try the procedure I described (push "reset" and "load" together then release "reset" then wait 1 second and release "load"), you should enter bootloader mode. The board should be recognized as PIC18F4550 Family device if I remember wright and the led should be blinking at 1 Hz.
If you have this behavior, the bootloader should work and I'll give you the procedure to download program and fuses using bootloader.

Which operating system does your friend use ?
Did he install the driver for the colorimeter ?

If you just plug in the usb connector, it should be recognised as "HCFR colorimeter", but I think the fact that fuses are not correctly programmed could lead to strange behavior.
Then if you try the procedure I described (push "reset" and "load" together then release "reset" then wait 1 second and release "load"), you should enter bootloader mode. The board should be recognized as PIC18F4550 Family device if I remember wright and the led should be blinking at 1 Hz.
If you have this behavior, the bootloader should work and I'll give you the procedure to download program and fuses using bootloader.

Which operating system does your friend use ?
Did he install the driver for the colorimeter ?He has Windows 2000 and I have XP.
Please go ahead with the bootloader procedure. If I get it right, this should reflash the chip correctly?

Did he install the driver for the colorimeter ?Yes, at some point it appeared at the Device manager; initially it wasn't installed correctly (had the exclamation mark), but finally he managed to install it.
He said that it appeared and disappeared sporadically, though. Maybe this has got something to do with the bootloading sequence.

He has Windows 2000 and I have XP.
Please go ahead with the bootloader procedure. If I get it right, this should reflash the chip correctly?
Yes, but then you won't be able to use low voltage programming any longer on this PIC.
I send it to you but it is in french.

Yes, at some point it appeared at the Device manager; initially it wasn't installed correctly (had the exclamation mark), but finally he managed to install it.
He said that it appeared and disappeared sporadically, though. Maybe this has got something to do with the bootloading sequence.

This sounds strange to me.
Once the driver is installed, the behaviour should be stable. If the probe is not detected, try to push and release "reset" push.

Thanks Benoit for looking into this. They also have a kit version that uses the PIC18F2550 (http://www.sparkfun.com/commerce/product_info.php?products_id=8265), but it's only $5 cheaper. Your team may even want to contact Sparkfun about adding the color sensor to the board, as they are very open to distributing project boards like this.

--Dan

Hi.
I began to deal with it.
I've modified one of my boards to fit with UBW pining and after modifying firmware, finally get it work (not so simple :rolleyes:).
I still have to adapt IR emitting function, but the color measurement function works.
I would like to get a board from sparkfun to test it with real environment, PIC 18F2455 and 20 MHz quartz but it stills out of stock.
I think I will buy the kit version but then I won't be able to test on 18F2455 PIC.

Hi.
I began to deal with it.
I've modified one of my boards to fit with UBW pining and after modifying firmware, finally get it work (not so simple :rolleyes:).
I still have to adapt IR emitting function, but the color measurement function works.
I would like to get a board from sparkfun to test it with real environment, PIC 18F2455 and 20 MHz quartz but it stills out of stock.
I think I will buy the kit version but then I won't be able to test on 18F2455 PIC.Thank you! These boards are extremely popular, and each batch may be sold out from backorders. The kit version sold out in 1 week, they got in 30 more, and those are selling fast. I will backorder one of the 18F2455 modules and see how long it takes.

--Dan

On the the 18F2455 module, you will have to remove the power supply LED because it is not possible to turn it off and it will be a problem for color measurement.

I've just ordered one ... :p

Yes, they're in stock again. My backorder should be on its way! :)

But now the color sensors are out of stock at Mouser until April 7... :(

--Dan

My order has been shipped yesterday ... just have to wait (it will give me time to prepare additional sensor board).

Got my 18F2455 module today. Wow this thing is tiny. Now I just need a color sensor and firmware... ;)

--Dan

I plan on building one as well, just waiting on Mouser. I will use the "point to point" solder on a proto-board method. I have a question. I noticed that the IR led is on the same side as the sensor. I thought that the IR led was to control a DVD player. If I just want to take readings manually, do I need to use an IR led at all?

I plan on building one as well, just waiting on Mouser. I will use the "point to point" solder on a proto-board method. I have a question. I noticed that the IR led is on the same side as the sensor. I thought that the IR led was to control a DVD player. If I just want to take readings manually, do I need to use an IR led at all?

No you don't need IR LED if you don't want to control a DVd player.
You can build the colorimeter without it.

Got my 18F2455 module today. Wow this thing is tiny. Now I just need a color sensor and firmware... ;)

--Dan
Good new.
I fear it will take longer to France ;)

Can you read the content of your PIC using Microchip bootloader and send the result to me so that I can check it before I receive mine ?

Here's what it read over USB. I assume this includes both the bootloader and the rs232 emulator demo code that was pre-installed.

--Dan

No you don't need IR LED if you don't want to control a DVd player.
You can build the colorimeter without it.

Thanks for that info. While waiting for Mouser, I re-read this whole thread carefully and found another problem. I had planned on using a DIY low voltage programmer (ART2003 PIC18Fxxxx) and WinPic800. Looks like I need another DIY programmer. I can't justify buying a MicroChip programmer for just a few uses.

the other changes I made was to move the photo diode arrays closer together and rotate one 90 deg.

What are the advantages to moving them closer and rotating?

What are the advantages to moving them closer and rotating?

I’m not sure there’s any advantage (or disadvantage) but there is some small clocking sensitivity on some displays so having them clocked 90deg relative to each other gives you one of each without rotation the probe. I did it because it just seemed totally redundant otherwise…

Dave

OK, I thought it may have something to do with the characteristics of polarized light coming for certain displays.

OK, I thought it may have something to do with the characteristics of polarized light coming for certain displays.

As I noted, there has been some clocking sensitivity reported. For many meters not just the TAOS array. P/S polarizers seems a good cause/effect but it's also been reported from plasma and CRT so maybe there's more to it. One user reported differences with a D2/plasma depending on how the probe was oriented to the screen.

When I carefully tested for it, I can't say I saw any effect.

http://www.avsforum.com/avs-vb/showthread.php?p=10514411#post10514411

Dave

Thanks for that info. While waiting for Mouser, I re-read this whole thread carefully and found another problem. I had planned on using a DIY low voltage programmer (ART2003 PIC18Fxxxx) and WinPic800. Looks like I need another DIY programmer. I can't justify buying a MicroChip programmer for just a few uses.

Yes it won't work with low voltage programming.
My programmer is this one
http://fribotte.free.fr/bdtech/programmateur/progPic.html
It think this one is the same :
http://www.sparkfun.com/commerce/product_info.php?products_id=8#
Another solution is to slightly modify the pcb so that it works with LVP but. you have to cut one track, put a pull down on one and the other one to +.
I think it should work but have never tested it.


What are the advantages to moving them closer and rotating?

We put 2 sensors because we hoped that by adding a filter on one, we could work without calibration file but we haven't found a solution yet.

Ki_, Thanks for the links, the JDM2 seems easy enough to build. I think I even have all the parts!

Ki_, Thanks for the links, the JDM2 seems easy enough to build. I think I even have all the parts!

Be careful, the db25 is a serial port, not a parallel.

Here's what it read over USB. I assume this includes both the bootloader and the rs232 emulator demo code that was pre-installed.

--Dan

Thank's.
Yes there should be both bootloader and UBW firmware.
I wanted to check fuses state, I have them now ;)

Hi
I received mine this morning :p
I didn't think it was so small !!!
First tests seems ok for color measurement but I still have problem for IR driving.
I still have few more tests to perform.

Hi
I've fixed my IR driving problems and finished tests. Everything seems ok :)

Still have to do pining schematic.

Hi
I've fixed my IR driving problems and finished tests. Everything seems ok :)

Still have to do pining schematic.Great!

Hopefully I'll get the color sensor later this week. I plan to solder it to this SOIC-8 breakout board (http://www.sparkfun.com/commerce/product_info.php?products_id=494) I ordered with my PIC module.

--Dan

Bad news guys, I just emailed Mouser and the TCS230Ds won't be in 'till 5.2.08 :mad:

The price will be $6.69EA.

Bad news guys, I just emailed Mouser and the TCS230Ds won't be in 'till 5.2.08 :mad:

The price will be $6.69EA.Hope you're right, I canceled with Mouser and ordered from Newark which has them in stock, but they're in the UK and will cost about $10 more because of the handling charge. Should be here next week.

--Dan

Maybe I'll do the same, do you have a link?

Maybe I'll do the same, do you have a link?This is my favorite search engine for electronic parts (http://www.findchips.com/). Just type in tcs230 to see everywhere you can buy it.

--Dan

Got my color sensors today from Newark!

Benoit, which pin connections are different from the original HCFR schematic? If I have time, I might finish it this weekend. Thanks.

--Dan

Hi
The pining is the following :
pin 1 : Vdd
pin 2 : Vdd
pin 3 :RB6 for sensor 0, RB7 for sensor 1
pin 4 : Gnd
pin 5 : Vdd
pin 6 : RB3
pin 7 : RB4
pin 8 : RB5
Add a 100nF capacitor between pin 4 and 5

IR driving : RB0

Done, passed the smoke test! :)
Just need new firmware...

--Dan

Cool!, I decided to wait for Mouser for my order, I also needed the PIC as well. Did the sensor come in a moisture-barrier bag? If not, did you bake them? This is what I found in the spec pdf...


Moisture Sensitivity
Optical characteristics of the device can be adversely affected during the soldering process by the release and vaporization of moisture that has been previously absorbed
into the package molding compound. To prevent these adverse conditions, all devices shipped in carrier tape have been pre-baked and shipped in a sealed moisture-barrier bag. No further action is necessary if these devices are processed through solder reflow within 24 hours of the seal being broken on the moisture-barrier bag. However, for all devices shipped in tubes or if the seal on the moisture barrier bag has been broken for 24 hours or longer, it is recommended that the following procedures be used to ensure the package molding compound contains the smallest amount of absorbed moisture possible.

For devices shipped in tubes:
1. Remove devices from tubes
2. Bake devices for 4 hours, at 90°C
3. After cooling, load devices back into tubes
4. Perform solder reflow within 24 hours after bake Bake only a quantity of devices that can be processed through solder reflow in 24 hours. Devices can be re-baked for 4 hours, at 90°C for a cumulative total of 12 hours (3 bakes for 4 hours at 90°C).

For devices shipped in carrier tape:
1. Bake devices for 4 hours, at 90°C in the tape
2. Perform solder reflow within 24 hours after bake Bake only a quantity of devices that can be processed through solder reflow in 24 hours. Devices can be re−baked for 4 hours in tape, at 90°C for a cumulative total of 12 hours (3 bakes for 4 hours at 90°C).

The sensors came in a tube inside a standard anti-static bag. I hand soldered each pin quickly (I only wish I still had access to a reflow oven), so I doubt the die got hot enough for damage. I think that warning is there because unlike most chips, the die is exposed, and reflow would get the die hot enough for physical damage from expanding steam or could permanently diffuse H2O into the die surface.

--Dan

I emailed Mouser wondering when they would receive stock of the TCS-230 and this is what I got...

"The manufacturer is having problems. As of now they have no idea when they can have that part made. Sorry - we have no control over issues like that."

:mad::mad::mad:

Here's some results comparing my HCFR probe against a DTP94 and Spyder2: Link (http://www.avsforum.com/avs-vb/showpost.php?p=13838027&postcount=6510)

It is much more accurate than the Spyder2. It tracks the DTP94 well from 10-100 IRE (not shown) as well as primary and secondary colors. But getting this accuracy requires creating your own calibration file against a known good reference probe, which is not hard at all. The Spyder2 is so slow compared to the HCFR and DTP94 that I can't stand to use it.

Too bad these chips aren't available anywhere right now. Newark had 500 in stock when I placed my order, but they were gone within a week. Someone is snapping up large quantities of these.

--Dan

dhnjp1: can you do a writeup on how to connect the TCS230/SOIC->DIP to the board from SparkFun? I believe thats what you did, right?


Also, where did you get the firmware for that to work?

Second, do you have any extra TCSs? Mouser is still out, and I'd rather not purchase from overseas. (Or would you be willing to put another probe together for purchase?)


tia

Hi,
You can find the pining in my post here http://www.avsforum.com/avs-vb/showthread.php?p=13673207#post13673207

For the firmware, I will make an official release soon ;)

dhnjp1: can you do a writeup on how to connect the TCS230/SOIC->DIP to the board from SparkFun? I believe thats what you did, right?

Also, where did you get the firmware for that to work?

Second, do you have any extra TCSs? Mouser is still out, and I'd rather not purchase from overseas. (Or would you be willing to put another probe together for purchase?)Here's a good beginner tutorial (http://www.kevinro.com/newdocs/learningcenter/surface.htm) with a video that makes it look easy. I think he uses too much flux, but that technique works well. If a joint looks too "dry" after soldering, I like to put an extremely small amount of solder on the iron tip, then quickly touch it to the pad/pin and the flux will make it flow over the joint.

I only ordered one extra chip for a spare. Sorry, not for sale.

--Dan

Newark has 700+ of these chips in stock right now. The $15 handling fee includes the shipping cost, so it's not that bad.

--Dan

Too late, they're all gone now. Wonder which company is snapping these up?

--Dan

Thanks. I just decided to splurge and get an eyeone something something.

http://www.amazon.com/gp/product/B000NRODT4******s9subs_c2_img1-2871_p?pf_rd_m=ATVPDKIKX0DER&pf_rd_s=center-2&pf_rd_r=1FGVVF3DW397KSRHR3KR&pf_rd_t=101&pf_rd_p=278240301&pf_rd_i=507846

Found one for 115, and i should get a 30$ credit from my amazon visa..so i think total was 85 ish.

Looks like the sensors are back in stock everywhere. Mouser has 566 and Newark has 764 in stock right now. Hope everyone waiting got there's.

--Dan

I first saw them on Newark, so I ordered some. A few days later I saw them at Mouser. Now I have to wait 10 to 15 days for shipping from Newark as apposed to 2 to 4 days from Mouser. Oh well, at least they will come in.

What is the brightest light source that this sensor can handle directly? Could I use it to, say measure the direct output of a bulb? The sun? :-)

This isn't a theoretical question; I'd like to start measuring/controlling the ambient light in various rooms.

Thanks!

I believe where it saturates is somewhat programmable (different ranges) and will depend on the matrix coefficients of a given cal file. I recently used mine to gauge the output form a high intensity LED array. I believed the response should have been largely linear but the response from the probe showed otherwise. It was flattening out (but not flat) when my readings approached 600 nits. I placed a ND2 filter inline and this knocked the amplitude by about an order of magnitude and then the response was largely linear as I expected.

If you were going to try to measure relatively high levels of illumination, my suggestion would be to try ND filters as an optical attenuator for relative rather than absolute measurements.

Dave

Hi.
It is not possible to program the saturation level of the sensor. The programmable output frequency range activates a frequency divider at the output of the sensor to reduce the maximum output frequency. It reduces measurement accuracy but dose not change saturation level except if frequency measurement device is not fast enough.
We uses full frequency range of the sensor (0 - 600 kHz) to get best possible accuracy. The PIC we use is fast enough to measure such frequency and then ther is no saturation due to measurement device.

You can find the saturation irradiance of the sensor in the TCS230 datasheet. It depends on the light frequency.

There was talk of a firmware change so that the sparkfun PIC18F2455 module could be used. Did that firmware ever get released? Where might I find it?

I finally got my probe together. It seems to work well. The only thing I have left to do is a configuration file. I have a Sony SXRD TV and thought I saw a configuration file for it, but for the life of me, I can't remember where it was. Anyone?

The Sparkfun module now uses the PIC18F2553 at 24MHz. The older version used the 18F2455 at 20MHz.

Does HCFR firmware v5.40 include support for the 18F2455 (20MHz) probe? Any future support for these boards, or is it a dead end?

--Dan

The Sparkfun module now uses the PIC18F2553 at 24MHz. The older version used the 18F2455 at 20MHz.

Does HCFR firmware v5.40 include support for the 18F2455 (20MHz) probe? Any future support for these boards, or is it a dead end?
Guess it's a dead end. :(

That's sad because this was the easiest way to build a HCFR probe. No custom pc boards, and 98% of the soldering already done for you.

--Dan

Hi,
I decided to build on of these HCFRs since the retail sensors costs so much. I can't find the original TAOS TCS230 sensors anywhere but Mouser does have some 856-TCS3200D-TR tricolor sensors. Does anyone know if these will work? Are they pin-to-pin compatible? I wasn't able to find the data sheet on the original TCS230.

Thanks for the help.... BTW, I'm building the version as per the French site. I do not have a fancy PCB like some of y'all were using.

Hope it works!

Just wondered, how do these DIY probes compare to, say, an Eye one LT, or Eye one pro? How do you calibrate the meters? Sorry if this has already been brought up.

Just wondered, how do these DIY probes compare to, say, an Eye one LT, or Eye one pro? How do you calibrate the meters? Sorry if this has already been brought up.

http://www.avsforum.com/avs-vb/showthread.php?p=10477985#post10477985

Hey guys, is there a compatible replacement for TCS230D since it is almost impossible to find it nowadays? I've found it only in one place and it costs 60$ !?

Hey guys, is there a compatible replacement for TCS230D since it is almost impossible to find it nowadays? I've found it only in one place and it costs 60$ !?

I have a few extra PCB's I had made up, i have all the parts to using the new Taos chip and it works great. I actually use the HCFR probe and HCFR to
calibrate my low end 10 and 20. I use Calman v4 and my i1 pro for everything else.

I can build one for you.

Athanasios

Thanks but I forgot to mention that I am in Europe so postage and possible tax (which you never know if you are gonna pay or not in advance) would cost me to much :(

Thanks but I forgot to mention that I am in Europe so postage and possible tax (which you never know if you are gonna pay or not in advance) would cost me to much :(

Where in Europe are you? I might be making a trip this winter to greece.
It might be cheaper to ship from there.

Athanasios

Serbia, but the problem could be our customs. If they decide I should pay taxes it could cost me over 50EUR easily.