Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

can not program 18F4550 with te20 & winpic (for ICD2 clo

Status
Not open for further replies.

alunaro

Member level 5
Joined
Jan 15, 2002
Messages
91
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Location
Another world
Activity points
770
winpic icd2

Hello,

I'm trying to program a 18F4550 (for ICD2 clone). I'm using the only programmer i have: a Te20-se. It's attached to a breadboard where is the 40 pin socket for pic (used the breadboard like adaptor)

I use Winpic (last version downloaded from site), configured to use te20.

I have tested with 16F84, 876, 877... all ok (note: i tested all pics both te20 sockets and breadboard). Only 4550 doesn't work.

1. It's not detected sometimes... All points to a bad connection in breadboard.

2. When winpic can detect it, i try to program.. but always report errors in different
addresses. Normally it fails in 1, 2 or 3%. 1 time it reached 6% :O lol xDD

10_1184972561.png


3. In hardware test Vpp is about 12.8V (winpic option).
When reading pic, it drops to 11.8-11.5 V.


4. Pics can be read and erased without problems.

Following some advises in other posts, i used some capacitor in PGC and GND, etc.
Still the same error.


Some idea?
 

winpic 18f4550

alunaro said:
Hello,

I'm trying to program a 18F4550 (for ICD2 clone). I'm using the only programmer i have: a Te20-se. It's attached to a breadboard where is the 40 pin socket for pic (used the breadboard like adaptor)

I use Winpic (last version downloaded from site), configured to use te20.

I have tested with 16F84, 876, 877... all ok (note: i tested all pics both te20 sockets and breadboard). Only 4550 doesn't work.

1. It's not detected sometimes... All points to a bad connection in breadboard.

2. When winpic can detect it, i try to program.. but always report errors in different
addresses. Normally it fails in 1, 2 or 3%. 1 time it reached 6% :O lol xDD

10_1184972561.png


3. In hardware test Vpp is about 12.8V (winpic option).
When reading pic, it drops to 11.8-11.5 V.


4. Pics can be read and erased without problems.

Following some advises in other posts, i used some capacitor in PGC and GND, etc.
Still the same error.

Some idea?

The most comun problem is in the cable length betwen your PIC and your Te20-se, use small length the most close possible.
Problem solved

Regards
 

winpic80

Another trick to try with the 18F range is 22pF capacitors from B7/PGD and B6/PGC to GND/Vss locally at the chip, i.e as close to the chip as possible and a 0.1uF cap from Vdd to GND/Vss.

You can also try adding a 10k pullup resistor to B7/PGD and Vdd.

Just some extra things to try
 

18f4550 trick with jdm

Thanks,

But nothing worked with 4550. Still the same error... only it looks a quite more stable.
I used all your suggestions.


I could program the 18F2550 without any capacitor or other trick. There are any difference with 18F4550 programming?
I can't undestand why can program the 2550 and not 4550. Only difference is that 4550 is in a breadboard attached to te20 with wires.... suspicious....


Thanks again.
 

winpic 2006 rs232

check for pins 23-25, I'm not sure the exact pin number,
but I think I read it somewhere that the pins need not be connected to anything during programming, the pins are usb's.

just a thought.
 

18f4550 not detected

I only have connected the basic pins: MCLR/VPP, PGD, PGC, VDD and VSS.
Rest are "float"


I spent the whole day (well, also yesterday) trying burn it... in different ways... No result.
 

18f4550 programming problem

PIC18F4550 have two power (Vdd,AVdd and Vss,AVss)
conect 100nF to Vdd and Vss

PIC18F4550 read an erase without problems ?
 

winpic te20

Yes, aparently read and delete without problems almost 100% of times (if detected).

I had problems to autodetect device, but... when i manually join the wire of vpp voltage to vpp pin in pic, winpic could autodetect. I suppose it was a bad connection due to bread board.. I solved this issue.

I tried with some capacitor in VDD-VSS (0.1uF suggested by foxabilo). "Programmer" looked more stable but still couldn't program the 4550.

Note: I'm using a rs232 wire of 75 cm. Should use one shotest? sure...

Well... let's go again xDD 100nF now xD I hope having one here (22:45h).

Thanks

Added after 31 minutes:

I attached the te20 directly to port.
No result with capacitor. As you can see it really detecs, reads and erase the pic.


Detecting:

26_1185225215.png


Reading:

0_1185225184.png


Erasing:

24_1185225267.png



All ok, less programming.


Note: when i write a "empty" buffer.. there is no error. I dunno what winpic does when programming.[/img]

Going to try with a transistor to increase VPP til 13V
 

te20 pic

try to program only 8 word rest all word 0xFFFF
if is ok need more power for program
add 100uF for power
 

    alunaro

    Points: 2
    Helpful Answer Positive Rating
program 18f4550

do you mean... create a empty buffer, modify the first 8 words manually and write?
 

winpic not detected

I tried with 8 bytes and was programmed perfectly... with full .hex the 100uF didn't work.
I'll check what's happening with vpp voltage...
 

jdm 18f2550 autodetect

Just a small point to make, 0.1uF and 100nF are the same thing :) also what PICS is suggesting is as additional 100uF (quite large) cap to Vdd/Vss this will allow a certain level of redundancy to the power supply, 100uF will run a PIC for a good few Milliseconds so its perfect for covering up poor power supply issues.

Also 85cm is too long, unless your talking uber quality armour shielded twisted pair cable. Personally I never go over 10cm for programming cable legnth. Twisted pair on all of the programming lines (one line data one line GND) will help greatly as the helical twist offsets the phase of regular interference and effectively removes it from the lines (mains hum etc) but nothing is better than prevention and that means short programming cable. Some people have reported good performance with old style 80 pin IDE cable with every other wire to GND, same with twisted pair RJ45/CAT5/CAT6 cable, thats the reason you see lots of RJ45/tgelephone type connectors on programmers.

To summarise, reduce the programming cable length as much as you can, and never over 1 foot/12 inches/30 cm long, add 100uF to the 0.1uF cap from Vss to Vdd and it should be fixed.
 

    alunaro

    Points: 2
    Helpful Answer Positive Rating
pic te20

This evening I'll try all suggestions... together. Til now all the efforts were unsuccessful.

Reduce lenght of programming cable (last days i attached programmer directly to port, but it's quite ackward -too many wires around xD-)
Use 0.1nF capcitor in VDD-VSS (i understood in VPP from PICS comment). I think I used it in my tests (i can't remember exactly no. It's 8:45 and my brain still is heating neurons)

Will I have to build other "low-cost" programmer by LPT port? Only time will tell... :/

Late will report my results.
 

icd2 winpic

Does not work.

I plugged the te20 in port directly, used a 100uF capacitor (well, 100u, 220u, 47u... and possible combinations...).

Program was written until 14-15%. Better than last 6%, but stil not enough.

:( :( :( :(


Late some tests using more capacitors (til 400uF or so in total), i could program until 48% xDD it looks like a championship.
 

can not program 18f4550 winpic

how long are the wires from the Te20 to the 4550 ?
 

icd2 clone 18f4550 erase

please post your schematic for programmer with your value for components
 

18f4550 breadboard

Hi

Thanks for your efforts..

Wires are less than 10 cm, 0.8 diameter.
Connecting capacitors to VPP doesn't change behaviour significatively.
In VDD-VSS it does.


I haven't the original schematic. I made this programmer long time ago, still works like first day -at least with 16F8xx series-.
It's a TE20-SE.

I found a schematic that is equal (i checked it). Its TE20-SEX:


https://obrazki.elektroda.pl/78_118...ki.elektroda.pl/78_1185398048_thumb.png[/img]

Only difference are transistors: NPN are BC547B, PNP is BC557 (in my programmer). Rest of componets are equal, with same values.
I hope it will be useful. Otherwise, i'll draw the schematic when have some free time.


Yes, this programmer has 2 LEDs. I haven't time today to remove them... to gain some few volts more :D
Therefore, tomorrow will try removing LEDs. It doesn't work, then try a simple and real JDM. It it doesn't work.. then... maybe try a EPIC. If not... pray xDD


PICS, winpic needs some special configuration to program your hex for 18f4550?

Thanks
 

winpic programmer

Hi

I just made it work !

Removing LEDs and using a 220uF capacitor, all solved.

I have tested with 2 PICS and no problem. None failed and read code after programming is ok.


Now... its time to test the icd2 programmer lol

Thanks !
 

Re: can not program 18F4550 with te20 & winpic (for ICD2

alunaro said:
This evening I'll try all suggestions... together. Til now all the efforts were unsuccessful.

Reduce lenght of programming cable (last days i attached programmer directly to port, but it's quite ackward -too many wires around xD-)
Use 0.1nF capcitor in VDD-VSS (i understood in VPP from PICS comment). I think I used it in my tests (i can't remember exactly no. It's 8:45 and my brain still is heating neurons)

Will I have to build other "low-cost" programmer by LPT port? Only time will tell... :/

Late will report my results.

Greetings,

Man, this is the solution: all these PIC Microcontrollers, have a PIN called PGM. For this PIC, the 18f4550, you must put this PIN to GND.

So, check the PIN PGM in the data sheet (i remember this PGM pin is the RB3 for the 18f4550, but check). Put this to GND (0V) and programming it with the Winpicprog, not the ic-prog (it doesn't work for me, yet).

Let me know.........

Good Luck.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top