TrippLite UPS Interface Problem
TrippLite Support has not responded to my email messages, so before contacting the TrippLite CPO, I am hoping someone out there on this Linux list has had success with the serial (DBD-9) interface from UPS to SuSE 7.3. With the ups monitoring software installed and TrippLite BC PRO 850 UPS connected, my system shuts down within seconds of entering: rcups start I am using a 9-wire cable that connects the same pins (straight-through) on either end. I have tried genericups type numbers of "4" and "5". Further, due to no answer from TrippLite Support, I attempted to do some reverse engineering. I disconnected the UPS from 110v, and monitored resistences and voltages on the unit's DBD-9 connector. I expect there are contact-closures on the other side. It ran my computer and CRT 40 minutes before the low-battery indicator lit up. Through the whole process, pins 3,6,7,8 indicated near-zero resistance together and connected to ground. No changes occurred through the power disconnect and discharge sequence! Your help appreciated. It seems like a good unit, and I would hate to have to return it to the vendor. ... Reed (reply to alta@alta-research.com) ----------email-to-TrippLite------------ Dear technical Support ... I am having difficulty getting your BC PRO 850 to work with Linux. Can you send me pin-outs for the the DBD-9 connector on the BC PRO 850? Or, can you tell me where to find the information? In other words, I need details for the "contact closure" style interface -- which pins are used for what event? I have spent a number of hours attempting to get the unit to work with a Linux system, without success. The BC PRO 850 manual contains no informantion on the DBD-9 interface. I have searched the TrippLite web site, without success. I have attempted to locate the active power-fail pins by taking ohm and voltage readings... The technical information would be a big help. No RS-232 cable was supplied with the unit. I am using a straight-through cable, where each pin is connected to the same pin at the other end. Sincerely ... Reed -- Reed White - ALTA RESEARCH - www.alta-research.com Phone: 877-360-2582 - Email: alta@alta-research.com
On Thu, 2002-02-14 at 10:45, alta wrote:
I am using a 9-wire cable that connects the same pins (straight-through) on either end.
FWIW - I grabbed a 9-pin to 9-pin serial cable from a huge "Box 'o Cables" - origin unknown - but molded on one end is "PC" and the other says "UPS". Presumably, they were for connecting some UPS to a computer. I have no idea of brand, but: **They were crossover cables**. At least, pins 2 and 3 were crossed over (that's all I cared about for my application. Actually, I didn't want a crossover, so I stuck two of these cables in series, and got my desired straight-through extension :-) ). -Gord -- Gordon Pritchard, P.Eng., Member IEEE Technical University of B.C. - Research Lab Engineer mailto:gordon.pritchard@techbc.ca direct phone: 604-586-6186
On Thursday 14 February 2002 13:45, alta wrote, in part:
Your help appreciated. It seems like a good unit, and I would hate to have to return it to the vendor.
Google searching on "tripplite ups pinout" reveals http://www.lugor.org/ups/genpower.txt. That should do the job for you. Bonus: The site is for Linux users. -- Regards, Malcolm KMail l.3.1 -- KDE 2.2.2 -- SuSE Linux 7.3 Remove the dots to email me
http://www.lugor.org/ups/genpower.txt. That should do the job for you.
Yes -- good info. Thanks. Although at first glance, the shown serial port side does not seem to exactly match any of the genericups types supplied with the SuSE 7.3 system. But, I think I can fix that. Diagram for TrippLite BC PRO to Linux PC: (Fixed-width font helps in viewing the following diagram.) UPS SIDE (9 Pin) SERIAL PORT SIDE PIN 3 o-------------+------------------------o DCD [POWER FAIL] | +---+ | | resistor | | 10 kilo-Ohm | | +---+ | +------------------------o RTS | +---+ | | resistor | | 10 kilo-Ohm | | +---+ | PIN 6 o-------------+------------------------o CTS [LOW BATTERY] Pin 1 o--------------------------------------o GND [COMMON NEGATIVE/GND] +----------+ Resistor PIN 2 o---------------+ 100k Ohm +-----------o DTR [INVERTER SHUTDOWN POS] +----------+ PIN 4 o--------------------------------------o RI [INVERTER SHUTDOWN NEG]
Malcom ... According to info faxed me from TrippLite: pin desc 1 "gnd", actually -12v derived from computer's SD pin-3 This feeds the emitter side of two opto isolators. 5 AC Fail, collector side of opto isolator, pulled up to +12 by 10K res in cable connected to computer's DTR pin-4. Connected to computer's CTS pin-8. -12 volts means AC Fail. 6 Low Batt, collector side of opto isolator, pulled up to +12 by 10K res in cable connected to computer's DTR pin-4.. Connected to computer's DCD pin-1. -12 volts means Low Batt. 2 Inverter shutoff, controlled by +12 from computer's RTS pin-7. Diode between UPS and computer, anode to computer. +12 volts from computer causes inverter to shut down. 4 Inverter shutoff, connected to computer's sig gnd pin-5. genericups 5 puts -12 on SD, +12 on DTR, and -12 on RTS. This all makes sense to me (and is compatible with measurements) for genericups 5 except for one bad thing: If I simulate AC Fail and Low Batt by feeding -12 volts to computer pins 1 and 8, my computer does not shut down. The daemon is running, but no cigar. ... Reed, alta@alta-research.com
On Friday 15 February 2002 15:58, alta wrote: Great information. Good to know TrippLite responded.
genericups 5 puts -12 on SD, +12 on DTR, and -12 on RTS. This all makes sense to me (and is compatible with measurements) for genericups 5 except for one bad thing: If I simulate AC Fail and Low Batt by feeding -12 volts to computer pins 1 and 8, my computer does not shut down. The daemon is running, but no cigar.
Do you have the resistors between 4 and 1 and 4 and 8 when you are testing? I would think that they would have to be present for the software to detect a valid level change. RS-232 inputs can swing either way if they are simply open circuited. Applying the -12v to 1 and 8 might not actually be changing their states. -- Regards, Malcolm KMail l.3.1 -- KDE 2.2.2 -- SuSE Linux 7.3 Remove the dots to email me
I made an adapter gizmo with resistors and diode. Re-booted (just as a precaution), and simulated power-fail and low-battery. Still, my system won't shut down. I verified that the daemon is running with genericups 5. It does not seem right that upsc and upsct fail, as below: alta@alta1:~> upsc localhost Unable to get variable list - Receive timeout alta@alta1:~> upsct localhost Unable to connect to localhost - Connection failure: Connection refused Hmmm, I wonder why they would have trouble with localhost. Like port numbers not matched, or blocked. I'm not real knowledgable in this area. ... Reed, alta@alta-research.com On Friday 15 February 2002 13:05, you wrote:
On Friday 15 February 2002 15:58, alta wrote:
Great information. Good to know TrippLite responded.
genericups 5 puts -12 on SD, +12 on DTR, and -12 on RTS. This all makes sense to me (and is compatible with measurements) for genericups 5 except for one bad thing: If I simulate AC Fail and Low Batt by feeding -12 volts to computer pins 1 and 8, my computer does not shut down. The daemon is running, but no cigar.
Do you have the resistors between 4 and 1 and 4 and 8 when you are testing? I would think that they would have to be present for the software to detect a valid level change. RS-232 inputs can swing either way if they are simply open circuited. Applying the -12v to 1 and 8 might not actually be changing their states.
On Friday 15 February 2002 18:31, alta wrote:
I made an adapter gizmo with resistors and diode. Re-booted (just as a precaution), and simulated power-fail and low-battery. Still, my system won't shut down. I verified that the daemon is running with genericups 5.
It does not seem right that upsc and upsct fail, as below:
alta@alta1:~> upsc localhost Unable to get variable list - Receive timeout alta@alta1:~> upsct localhost Unable to connect to localhost - Connection failure: Connection refused
Hmmm, I wonder why they would have trouble with localhost. Like port numbers not matched, or blocked. I'm not real knowledgable in this area.
I'm too new to Linux to be of any help either. Maybe someone else on the list will see this and have some info. If you haven't done so, you might find something at http://www.exploits.org/nut/. There is a genericups FAQ there I believe. There are also some UPS mailing lists mentioned. -- Regards, Malcolm KMail l.3.1 -- KDE 2.2.2 -- SuSE Linux 7.3 Remove the dots to email me
A few months ago, I bought the notebook listed in the Subject. A few days ago, I decided to install Suse 7.1 onto it. Once I found a program that would free up some hard disk space (the unit runs XP HE and NTFS), I really didn't have any problems. The install went fine, but afterwards, I had a hard time getting an X server up and running. Suse's Sax programs didn't get it running and my notebook's manufacturer didn't supply all the necessary data. I didn web searches for the info and came up nil everytime. I ended up using an older Trident driver (the video card is a Trident CyberAlladin-T, which seems to have no support in Linux) and got it running but through a frame buffer device. I cracked open the Suse 7.1 Handbook (page 249, section 8.3.3) and it told me all I needed to know. I'd thought that that info wouldn't have applied to a notebook but I tried it anyways and now I'm running X fine, at 1024x768x16. I thought I'd shared that with the list since I had such a hard time finding info on the web...now it'll be archived! :o) Next, I have a question regarding the sound card in this unit and support in Linux. The sound card is an ALi M1535. The notebook manual states it's compatible with the Soundblaster Pro. Does this mean I can use the kernel module for the Soundblaster Pro while in Linux? ALSA isn't helping to configure this card...it detects it, but that's about it. So, I wiped out ALSA and will go with OSS, as I did with all my Suse boxes...ALSA is $h*t, IMO...I've NEVER gotten it to work, no matter the machine. I've tried to use ALSA in four different boxes with four different motherboards and 4 different sound cards, with ZERO results. The boxed kernel doesn't support the ALi M1535 sound card. I tried to build my own kernel and compile support into the homegrown kernel but for some reason, 'make xconfig' isn't working (and neither is 'make menuconfig' or 'make config'). Per the Suse Handbook, everything's in place (all the kernel-building packages, that is), but the 'make' commands still won't work. Strange... So, I put out a bit of info for the Suse archives for any potential Toshiba 1800 series owners and I'm also asking for a bit of help or experience with the above soundcard. Regards, Ron Sinclair @ -=[ http://www.wigglit.com ]=-
On Friday 15 February 2002 04:12 pm, Ron Sinclair wrote:
A few months ago, I bought the notebook listed in the Subject.
A few days ago, I decided to install Suse 7.1 onto it.
Once I found a program that would free up some hard disk space (the unit runs XP HE and NTFS), I really didn't have any problems.
The install went fine, but afterwards, I had a hard time getting an X server up and running. Suse's Sax programs didn't get it running and my notebook's manufacturer didn't supply all the necessary data. I didn web searches for the info and came up nil everytime. I ended up using an older Trident driver (the video card is a Trident CyberAlladin-T, which seems to have no support in Linux) and got it running but through a frame buffer device. I cracked open the Suse 7.1 Handbook (page 249, section 8.3.3) and it told me all I needed to know. I'd thought that that info wouldn't have applied to a notebook but I tried it anyways and now I'm running X fine, at 1024x768x16.
I thought I'd shared that with the list since I had such a hard time finding info on the web...now it'll be archived! :o)
Next, I have a question regarding the sound card in this unit and support in Linux.
The sound card is an ALi M1535. The notebook manual states it's compatible with the Soundblaster Pro. Does this mean I can use the kernel module for the Soundblaster Pro while in Linux?
ALSA isn't helping to configure this card...it detects it, but that's about it. So, I wiped out ALSA and will go with OSS, as I did with all my Suse boxes...ALSA is $h*t, IMO...I've NEVER gotten it to work, no matter the machine. I've tried to use ALSA in four different boxes with four different motherboards and 4 different sound cards, with ZERO results.
The boxed kernel doesn't support the ALi M1535 sound card. I tried to build my own kernel and compile support into the homegrown kernel but for some reason, 'make xconfig' isn't working (and neither is 'make menuconfig' or 'make config'). Per the Suse Handbook, everything's in place (all the kernel-building packages, that is), but the 'make' commands still won't work. Strange...
So, I put out a bit of info for the Suse archives for any potential Toshiba 1800 series owners and I'm also asking for a bit of help or experience with the above soundcard.
Regards,
Ron Sinclair @ -=[ http://www.wigglit.com ]=-
Have you looked at http://www.linux-on-laptops.com/toshiba.html ? -ronc
Have you looked at http://www.linux-on-laptops.com/toshiba.html ?
-ronc
Yes. That's one of the first places I found. Specifically, this one (which is a different page but linked at your above URL) -----> http://www.hartzler.net/satellite/ That's the only URL that's specific to my notebook. He used Redhat and he reported that he had no problem with sound. But Redhat is a totally different beast than Suse and I'm not about to put RH on any of my systems. :o) Regards, Ron Sinclair @ -=[ http://www.wigglit.com ]=-
On Friday 15 February 2002 04:32 pm, Ron Sinclair wrote:
Have you looked at http://www.linux-on-laptops.com/toshiba.html ?
-ronc
Yes. That's one of the first places I found. Specifically, this one (which is a different page but linked at your above URL) -----> http://www.hartzler.net/satellite/
That's the only URL that's specific to my notebook. He used Redhat and he reported that he had no problem with sound. But Redhat is a totally different beast than Suse and I'm not about to put RH on any of my systems.
:o)
Regards,
Ron Sinclair @ -=[ http://www.wigglit.com ]=-
Yeah, I saw that page. Remarkably unhelpful. :) I don't suppose that any of the other Toashiba's have similar sound hardware that someone has done a better job of documenting?
participants (5)
-
alta -
Gordon Pritchard -
M. Clark -
Ron Cordell -
Ron Sinclair