Sorry to revive this thread, but I have a question that relates to it and I am hoping to get an answer from some of the previous contributors...
Above is the DIYinHK repair 18w pcb
I have been staring at this thing for hours, reading different threads on different websites and flipping through the toshiba data sheet for their chip trying to find the answer to this question. in the toshiba data sheets (
http://www.toshiba.com/taec/components2/Datasheet_Sync/201112/DST_TB6588-TDE_EN_28124.pdf) the VSP trace is referred to as receiving a signal between 0 and 5 volts. If it gets 5 volts, which it receives when the bridge between vsp and vref is left intact, it will spin at 100%. That being said, and Maybe I have something completely wrong, but if the bridge between VSP and Vref is cut, can't VSP be wired up to a motherboards PWM fan header?
If a female pin was set up with v+/gnd/tach/vsp (vsp becomes our signal from the motherboard) from the pcb and plugged into the header, would it not function properly?
The Toshiba data sheet seems to confirm it, Page 7, point 6, titled "Motor Speed Control pin (vsp)"
Quote:
An analog voltage applied to the VSP pin is converted by a 7-bit AD converter and used to control the duty
cycle of the PWM.
0 ≤ VSP < VAD (L) → Duty cycle = 0 %
VAD (L) ≤ VSP ≤ VAD (H) → Figure on the right (1/128 to 127/128)
VAD (H) < VSP ≤ VREF → Duty cycle ≈ 100 % (127/128)
That seems a bit too simple, but I would love it to be true...the reason why I don't think it is that simple is because of Page 6, point 5, titled "PWM Frequency". It specifically mentions the use of an external resistor and capacity being used to determine the pwm frequency.
Quote:
The PWM frequency is determined by the value of the external capacitor and resistor, and the logic level of
the FPWM pin (which has a pull-down resistor).
FPWM: High : fPWM = fosc/128
FPWM: Low or Open : fPWM = fosc/256
The PWM frequency must be sufficiently high relative to the electrical frequency of the motor and within
the range permitted by the driver circuit.
The PWM turn the high-side output transistors off.
Another user on another forum mentions needing a resistor and cap as well...
http://www.xtremesystems.org/forums/showthread.php?264957-Laing-DDC-pump-repair-PCB-mod&p=4817069&viewfull=1#post4817069
He links to another data sheet referring to the use of a motherboard pwm header connected to this res/cap and then to the pwm device...
Quote:
The Image he posted
What is bothering me about this is that the data sheet he links to refer to using a res/cap and a gate...but say the output will be analog pwm.
So the question becomes, is this extra res/cap combo simply a line filter to clean up the pwm signal from the motherboard? In which case, can we do without these components and connect directly to a PWM motherboard header or appropriate pwm controller?
Combining all this information, I believe it to mean that if you want the frequency of the PWM signal from your controller to be different from its default frequency, you have to add the right combination of res and cap to get the frequency you would prefer. Otherwise the pwm will operate at that frequency. Meaning, as long as the frequency of the controller you are using is at an acceptable noise level, you can just hook it up straight, other wise add the right combo till you get the frequency you want.
