Can you share the document?The rectifier sends a lot informations via can bus and it also sends alarm codes if you ask for them.
I have a complete can protocoll from a Huawei compatible rectifier. I think most of the codes will work for the Huawei, but I have not checked it, as I had no need for alarm codes and no time to test things I do not need.
How do you connect the SMU02B to the recitifier?I have SMU02B. There are no fan settings in it.
You can only see the temperature of each unit.
Thank you for your input. Pulse doubler might come in really handy. In this case it doesn't seem to care about the rpm too much as long as it sees it spinning. I figured out what the issue is. The standard seems to be blue wire for control and yellow wire for sense, with the yellow wire beside v+. That's what I saw on Nidecs website when checking a very similar fan model and on google. In reality they used the yellow wire beside v+ for control and brown wire for sense.. It works now but I also had to use a resistor because even the much slower fan is loud and with a resistor it's tolerable. The issue isn't the decibels but the frequency of the sound, a normal computer fan will have low frequency sound so even at higher db its much more comfortable. Nobody that sees this try to place a resistor on the original fan, it needs a 1.5w resistor at least or the resistor will turn red hot and melt the surrounding wires.Best guess is the tachometer is being monitored and when it runs slow the red light flashes to tell you there's a fan problem.
If you don't need the cooling the fan provides at it's full speed, you could install a pulse doubler (or greater) that creates two pulses for every tach pulse that comes in. Then if the PSU is controlling the fan speed (presumably via controlling it's input voltage) it can turn the speed down on it's own.
If it doesn't do it on it's own you can then use the resistor to reduce it's speed.
Pulse doublers, or frequency doublers, can be made in a number of ways. A quick google finds this useful thread
Are you willing to share you (compatible) protocoll documentation?The rectifier sends a lot informations via can bus and it also sends alarm codes if you ask for them.
I have a complete can protocoll from a Huawei compatible rectifier. I think most of the codes will work for the Huawei, but I have not checked it, as I had no need for alarm codes and no time to test things I do not need.
From my experience having multiple psu's on the same can bus and the outputs connected in parallel resets the so called "offline mode" after a power loss. After setting a certain voltage or current during operation and writing it to their internal rom (trough canbus) they keep their preset voltage or current, etc. until they are power cycled. That means that all the psu's jump back to their default preset, which is about 54ish volts, aswell as no current limit. I think that this must be a safety feauture by huawei because when one psu for example has a fault on the canbus and is therefor set to a different voltage than the others, a voltage overshoot or even a high curtrentflow between the units can be prevented. Not sure about this tho. Still, just a quick can message after the reboot will set all the units back to the original parameters as desired. Furtmore there is the following command which allows to send ONE or Multiple psu's to sleep. The ID of the frame: 0x108080FE (all psu's on the can bus) or, 0x108180FE for PSU1, 0x108280FE for PSU2, 0x108380FE for PSU3 and so on. Finally the actual Data of the can Frame: send following for to turn ON: 01 32 00 00 00 00 00 00, aswell as the following to turn OFF: 01 32 00 01 00 00 00 00.Would love to see that as well! Just got hands on 4850G2 and having issues for some reason, maybe someone can give a hint?
I boot it with both sense pins connected to ground. The unit boots well, no red/yellow lights and I’m able to communicate with it over can. Able to read data as well as request voltages. The issue I’m having is this:
1) I request 55v with byte[1] being 0 and get instantly to the 55v output. -all good
2) I request 55v with that byte being 1, and of course no change. I unplug power from the ps, wait 30 seconds and power it back on. Still, bolt voltage is like 48v or something around there. -that’s not what I expected.
3) I do the same command as I did on number 2, though let’s say set it to 53v, but this time abort all communications with the PS for 60 seconds. (Don’t send any more messages). The yellow light starts to blink indicating PS went offline mode and Voltage drops down to 53v. Thats expected and correct. I reboot the PS, and voltage is back to damn 48v!
Also, just with fresh boot, only command being sent is to request data every 10 seconds, my fan is spinning on max with idle AC draw of 23watts! Is there any way to use can to put this unit offline until I need it?
The following should help you out.Are you willing to share you (compatible) protocoll documentation?
The Input current limit and on off parameter I can not find any where.
one page backAre you willing to share you (compatible) protocoll documentation?
The Input current limit and on off parameter I can not find any where.
Yep already tried that command. Sending 0x01 3rd byte turns off the fan and cuts output, but unit is still pulling over 25watts. So not sure if there is a way to actually truelly put it to sleepFrom my experience having multiple psu's on the same can bus and the outputs connected in parallel resets the so called "offline mode" after a power loss. After setting a certain voltage or current during operation and writing it to their internal rom (trough canbus) they keep their preset voltage or current, etc. until they are power cycled. That means that all the psu's jump back to their default preset, which is about 54ish volts, aswell as no current limit. I think that this must be a safety feauture by huawei because when one psu for example has a fault on the canbus and is therefor set to a different voltage than the others, a voltage overshoot or even a high curtrentflow between the units can be prevented. Not sure about this tho. Still, just a quick can message after the reboot will set all the units back to the original parameters as desired. Furtmore there is the following command which allows to send ONE or Multiple psu's to sleep. The ID of the frame: 0x108080FE (all psu's on the can bus) or, 0x108180FE for PSU1, 0x108280FE for PSU2, 0x108380FE for PSU3 and so on. Finally the actual Data of the can Frame: send following for to turn ON: 01 32 00 00 00 00 00 00, aswell as the following to turn OFF: 01 32 00 01 00 00 00 00.
Greetings
Julian
Yeah that is a little above my knowledge. Yes output is 0, but energy is going somewhere. I looked at the temperature data and when set to power off mode using that command, I do see both intake and exhaust temperatures raise slightly. Also I can hear electrical noise come from the supply, so my assumption is that it just cures output and fan, but energy is still being used and converted to heat clearly, which seems to be much more then 7w. Again, could be wrong but that was just my observation.25w apperent power.
Power factor at 230Vac is <0.1 at 0W output power.
<7W standby power at sleep mode