Project description
Development
Mailing List
Contact
|
Project description
Development
Mailing List
Contact
|
AID: Problems on the 8 bit bus
From: mike2004 mike2004
acc.ocad.on.ca
Date: Fri, 19 Nov 2004 10:13:59 -0800
: Sent via the Art Interface Device mailing list: aid@interaccess.org : Use your "Reply All" to reply to the list, "Reply" for private response I did a bunch of tests on the AID after everyone left last Sunday. Since the InterAccess programmer doesn't work this week even on the laptop where it did work last time we met I am still using the version 1.4 of the firmware hex file. power up shows - Hello $Revision: 1.4 $ main.c $Revision: 1.4 $ aid_cmd.c $Revision: 1.5 $ aid_lib.c Ready - with nothing connected regulator does not heat up - with aid_output_board connected at address 1 and Mel's led board connected to the 10 pin header all leds light up briefly at power up but then go off immediately, after 10 minutes the regulator is barely warm. with aid_ip_board connected at address 0 and aid_output_board connected at any address and Mel's led board connected to the 10 pin header leds at bit 0 to 7 up and stay on for about 3 seconds at power up, after 10 minutes the regulator is cool. - with aid_ip_board connected at any address other than 0 and aid_output_board connected at address 0 and Mel's led board connected to the 10 pin header leds at bit 0 to 7 up and stay on for less than a second at power up, after 10 minutes the regulator is cool - with aid_ip_board connected at address 1 and aid_output_board connected at address 1 and Mel's led board connected to the 10 pin header leds at bit 0 to 6 light up and stay on indefinitely while 7 stays on for 3 seconds only at power up, after 10 minutes the regulator is very hot. - with aid_ip_board connected at address 1 and aid_output_board connected at address 2 and Mel's led board connected to the 10 pin header leds at bit none of the leds light at power up, after 10 minutes the regulator is warm - the last test was consistent until I went to another address at which point it became random whether the leds would do a 3 second flash or not. Also random as to how many leds light up. The leds seem to turn off at the same time as the startup message gets printed to the screen. - with aid_ip_board connected at address 1 and aid_op_board connected at address 1 and Mel's led board connected to the 10 pin header all leds light at power up but bit 7 goes out after 3 seconds with bits 0 to 6 of the leds remaining on, after 10 minutes the regulator is very warm. This is not completely consistent and some time fewer lights will come on and stay on in different combinations. - with no aid_ip_board connected and aid_op_board connected at address 1 to 16 and Mel's led board connected to the 10 pin header it is random how many leds will light up briefly at power up but then go off either immediately or after 3 seconds, after 10 minutes the regulator is cool if the leds go out. - with no aid_ip_board connected and aid_op_board connected at address 1 and Mel's led board connected to the 10 pin header , after 10 minutes the regulator is barely warm. - with no aid_ip_board connected and aid_op_board connected at address 1 and Mel's led board connected to the 10 pin header- - if "w 1 255" is written and all leds are turned on, after 2 minutes the regulator is very hot. After 3 it is too hot to hold onto. - Added aid_ip-board at address 0 to 15 excluding 1: - start up at each address would light a random number of leds tending towards all of them for 3 seconds. - writing functions normally with regulator heat relating to number of leds lit. - with aid_ip_board and aid_op_board at any same address a random number of leds tending towards 8 will light - with a few leds turning off after 3 seconds and the rest remaining on as notes above, the first write command gives unpredictable incorrect results but subsequent commands will give the correct results. Even if only one or even none of the leds are set to on by a second write command the regulator is too hot to touch. Observations: It is not desirable that the leds light up at power up. What is the technical explanation and how can we fix it? It seems having ip and op boards at the same address is dangerous. Bits get stuck in on position. What is the technical explanation and how can we fix it? Resistors of 220 ohms for the small leds are probably not high enough if they are going to be run from the main board power supply. I have seen values as high at 10k used on similar circuits. What values should we agree on? Mike ps I can confirm what Mel found - that aid_op_3 has a stuck 7th bit. The 573 latch is the suspect. It should be replaced this Sunday. : messages saved at http://www.interaccess.org/aid/list : unsubscribe/help requests to mailto:Majordomo@interaccess.org