HP HSTNS-PL11 psu Over Voltage Protection voltage increase?

[iz1fks]

ok, I removed any mod and with 12V ouput, ADC voltage is 3.1V. Installed the new voltage mod according to reply#8 by ubrain user, and the 22k resistor from adc pin 24 to ground and with 3.3V @13.8V ouput no OVP issue varing voltage up to 14.54V (max available).
Don't have power resistor to perform test with high current but I will test on field with my transceivers requiring more than 20A (yes, you're right I'm an ham!), eventually changing the resistor to raise the OVP, but quite sure that it's ok yet.

About the rf noise, it depends on several fators, equipments, frequencies used etc. usually a common mode filter wounded on ferrite on the ouput could help with conducted emi in hf or low vhf range. A ripple up to 100mV is easily visible with oscillocope on the output and also high frequency noise unfortunately; this is a drawback of these items, but I have not understood if it's due to a poor filters to reduce ripple and hf nosie (not need for application they are designed for) or due to a circuit design topology. I have seen similar issue on my dps-600 power supply.

I attach a couple of images to help any PL12 owner that need to perferm the mod (of course I don't take any responsability for the mod, people need to know that these power supplies can be very dangerous items!)

I would like to use a quieter fan, anybody know the pinout of the 4 wire pwm connector?

[eblc1388]

I would like to use a quieter fan, anybody know the pinout of the 4 wire pwm connector?

The industry standard for PC PWM controlled fans all follows the same connections, though the color of the wires might be different. These 4-wire fans will also run at full speed in a 3-pin connection. Starting from pin1 to pin 4 are:
Gnd, +12V, tacho feedback, PWM control input. It would be out of the ordinary that those fans on the power supplies would use difference connections but check to be sure.

 

[eblc1388]

Hello. Thats probably my first post.....
In the picture I present an example of a control loop for the power supply.

Hi, Jerycho. Have got the remaining pages of the power supply you have posted in your first post, i.e. the PS-3601-1F? If so, can you upload a copy to the forum?

I visited the various links you have provided but unfortunately it only allow visitor(me) to view the first two pages of that particular schematic.

Thanks.

[Jerycho]

Of course. I found a method to get the first 30 pages of documents on ishare.iask.sina.com.cn, but I need some time to build the index as my method requires getting an image from ishare.iask.sina.com.cn and splitting the image file by the address of the pattern found in image. The link includes an example file representing the first 30 pages of the document. You can open the file in your favorite hex editor and see what I mean by searching for "JFIF". I think by the end of this week I will be able to deliver about 10 power supply schematics worth your attention.

https://mega.nz/file/v1AixLpT#Pj421U9LQQVkBPhbi0hopAWHadVfBRhdHok4hvij0K0

[eblc1388]

Thanks for your effort.

I downloaded the file from mega.nz and examined it in a HEX editor. My first impression is the file is a collection of many jpeg files concatenated together into a large file. If we can find out the starting address of each individual file, then we can extract the individual jpeg files and then convert them back into a pdf.

Interesting.

p.s. Success. I search the net a bit and find out a way to automatically split out the individual jpeg files, using the awk command under Linux. Just create a temporary directory and place the merged jpeg file into it.  Then execute the command with the correct filename to match with the merged filename. I got all 50 jpeg files. Attached is the schematic in pdf format.

Code: [Select]

awk 'BEGIN {RS="\xFF\xD8\xFF\xE0\x00\x10\x4A\x46\x49\x46"; FILENUM=-1} {FILENUM++; if (FILENUM == 0) {next}; FILENAME="image0"FILENUM".jpg"; printf "%s",RS$0 > FILENAME;}' PS-3601-1F.jpg


[Jerycho]

So basically if you found something interesing you need to get link to image rendered on main page ie:

https://swf.ishare.down.sina.com.cn/avumxnYYL7B.jpg?ssig=0UZsQwnRAS&Expires=1637313011&KID=sina,ishare&range=0-63922

Then you need to remove last parametr from url, so you can get mostly all documents. Sometimes just svg format is available.
 Cheers

[eblc1388]

I tried downloading several documents and the method works. You rocks.

If you have Linux, have you tried the "awk" extraction method I mentioned above?

[Jerycho]

I have somewhere python script for spliting file into images. I'm currently not using linux systems. Btw I have done some dc simulation of control boards. In the simulation, affecting voltage references affects voltage/current limits. I beginning thinking about design some control board based on mcu and some dac (in my way is a pic24fj64gb002 (build in 1x 10bit adc muxed up to 14 channels) and mcp4527 12 bit dac's for voltage/current adjustment, but I'm currenlty studying many books from ti, analog about theory and calculations and have some time shortage.

[Jerycho]

Thanks @eblc1388 - I can confirm that works here on my 650EB!
Well, it works on one of them. The other has the relay click on and off and immediately goes into 'orange light' error mode. I suspect I may have damaged that one during my investigations. I think that one is destined to become 'spare parts' :-(
The one that does power up then spins its fans up pretty fast after a few seconds, and they seem to stay at that speed (much as they do when used in the server chassis), even with no load attached. Not the quietest PSU I think.
I'll go update my github docs now. Many thanks!

Hello. I found doc about server chassis which contains this psu.
https://www.intel.com/content/dam/support/us/en/documents/motherboards/server/chassis/sc5650/sb/e39531004_sc5650_tps_r1_2.pdf
Pinout is on page 45

[VA3HDL]

Hello everyone,

I have the HP HSTNS-PL11 psu made the two mods and everything is good for setting up the 13.8V needed for ham radio and not tripping the OVP.

However, I have two issues that I will need some direction on how to fix or adjust:

1. The OVP triggers at very high voltage for my use case - Close to 17v - How do I bring this down to 14.5v max under load?
2. I noticed that since I made the mode, the PSU fan goes really quiet and down in speed when the PSU is under load. Quite the opposite behavior before the mod. Any ideas?

Thank you,
Pablo

[eblc1388]

1. The OVP triggers at very high voltage for my use case - Close to 17v - How do I bring this down to 14.5v max under load?

If you have added the 13K resistor to increase the OVP trip point, then you can increase the value of this resistor to lower the tripping voltage level to the original trip point(by using a very high value resistor) or any level in between 17V and 13V. You can try 22K for a start to see if the unit operates correctly with 13.8V load.

2. I noticed that since I made the mode, the PSU fan goes really quiet and down in speed when the PSU is under load. Quite the opposite behavior before the mod. Any ideas?

Mine does that too but it has nothing to do with the modification. The fan speed will increase automatically when the internal temperature goes up.

[VA3HDL]

Thanks eblc1388

For #1

I will try with a variable resistor to find the right voltage spot, then switch to fix one.

For #2

The fan was working as it should before the mod, going up with load and temp and going down without any load. The issue is now when PS is under load the fan goes down, instead of going up as it was doing before.

Thanks.

[PA3VOS]

Hi all,

With a big THANKS to eblc1388 I modified the OVP on a 460W version that has a custom-made PIC called A4315-1. it is also a 28-pin IC. the PCB does look somewhat different from what I have seen here in this thread. I added a 33K resistor between pin 24 en GND and it worked for me perfectly!

73's PA3VOS

[eblc1388]

The power supply mentioned above is the 460W HP HSTNS-PL14.

Thanks PA3VOS for letting us know about the modification.

[PA3VOS]

Edit:
I had to increase the resistor value to 33K to ensure the OVP was increased to 14,5V.

[Ignatovd]

Hello to all!

have PL11 rev 16 and I can't get an OVP of 14.5V.
In this revision there is a PIC called A4315-1
Photo as in the post of 21.04.2022
Without load my power supply picks up the voltage to 14,4V , but as soon as I connect the load the voltage drops to 13,6V and changing the resistance has no effect on the voltage. With the load and at 13,6V my power supply makes a noise, but this noise comes from the board and not from the cooler.

- problem in this video

Maybe you have a solution for my PL11 rev16 power supply?

[eblc1388]

Without load my power supply picks up the voltage to 14,4V , but as soon as I connect the load the voltage drops to 13,6V and changing the resistance has no effect on the voltage.

It sounded like your power supply has entered output current limiting mode. How much current loading are trying to draw from it?

For a 1200W 12V power supply, it is rated at 100A max. As you have now increased the output voltage to 14.4V, the maximum current it can provide will be lower at 1200W/14.4V = 83.3A.

From the image of your supply label, it stated a max current of 160A which is wrong. It is unthinkable for a genuine big company like Lite-On to have made such a big mistake on the label of its product and not being discovered before the product has left the factory. This led me to suspect that the power unit is possibly a fake one and so had a even much lower output power.

 

[Jerycho]

I think you're using too long potentiometer cable, feeedback path is very fragile to noise. Try to use fixed resistor or make shorter you pot connection cables.

[hardocp]

Hi everyone,

with help from this thread, I managed to modify a Lite-on PS-2751-2F-LF to output 14.4V, the control board is very similar (maybe even identical). The OVP circuit is different but I found the right resistor to change to fix it - let me know if a picture of the mod would help anyone.
The issue that I have now is similar to the one reported by Ignatovd: over 11A @14.4V, the power supply cuts output to ~13.6V and starts buzzing. This behavior continued up to 23A and above probably - I did not have a bigger load to test with.

Any ideas on how to debug this further?

[Ignatovd]

Thanks for pointing out the incorrect specifications on the label!

The author in the comment under this video also suggested that it is a fake: "Another possibility is that you may have purchased a fake power supply. Genuine power supplies typically have serial numbers starting from 1, whereas yours starts from 3."


In this video, the power supply has all the refinements removed. With no load, the voltage is 12.4 volts. It operates at 75-76 amps at 12 volts. But when you try to increase the load over 80 amps, the protection kicks in and the power supply shuts down.
Conclusions:
- I did not damage the unit during modification, it works;
- the 1200 watt power indicated on the cover does not correspond to reality, so the unit is most likely fake.
And I'm still looking for a way to bypass the OVP to about 14.5 volts.

The issue that I have now is similar to the one reported by Ignatovd: over 11A @14.4V, the power supply cuts output to ~13.6V and starts buzzing. This behavior continued up to 23A and above probably - I did not have a bigger load to test with.

Glad to know that my problem is not an isolated one. Maybe you have found a solution in the intervening time?

[Ignatovd]

I think you're using too long potentiometer cable, feeedback path is very fragile to noise. Try to use fixed resistor or make shorter you pot connection cables.

I checked your position, soldered the permanent resistors and without using wires. Unfortunately, it continues to buzz at voltages above 13.5-13.6V under load.

[dimon_ua]

Good day.
Tried to replicate your mod using the PC817 on the PL11 unit. But somehow it turns out strangely. The output voltage is 3.72V.
It is worth disconnecting PC817 from the board, as the voltage again becomes 14.6V 
Tell me where I'm wrong.
Thank you!

[eblc1388]

Tell me where I'm wrong.

You have got the opto-coupler PC817 input/output side reversed.

I've reviewed the original picture of the mod and noted that there is some dirt on the underside of the opto-coupler, which you probably mistook as the round "dot" for pin #1. The schematic shows the correct connection though. I'm sorry for the unclear picture which led you to have connected the opto-coupler incorrectly. I've also updated the original picture to indicate clearly where pin#1 of the opto-couple is located.

[dimon_ua]

Thank you friend!
The scheme worked. It is possible to adjust the current starting from 50A and above.
Although honestly speaking, I thought there would be a wider range.
For example, adding a variable resistor in parallel with the 802Ω resistor, on the board near the left potentiometer, allows you to adjust the current starting from 30A. The lower the resistance, the lower the current. But even a short circuit gives a current reduction of at least 30A.
Therefore, it is not at all clear how in China they manage to adjust the current starting from 1A

[eblc1388]

Therefore, it is not at all clear how in China they manage to adjust the current starting from 1A

This modification is a simple maximum current limiter, and is not meant to be adjustable once trimmed to the required current value.

If you want a very wild range of adjustable constant current setting, you can refer to the method in Reply#41~42 of this thread kindly offered by forum member Jerycho. He did managed to set the current down from maximum to about 1A.