数字电位器多久才能回应？

因为它不是一个真正的雨刷必须移动，我会假设相同的公司或DEC命令。



      以下为原文

    Being it is not a real wiper that has to move, I would assume the same as a inc or dec command.

与SPI通信的速度相比，它应该非常快。



      以下为原文

    Compared to the speed of SPI communication it should be really quick.

我懂了。因此，一个10MHz的SPI时钟，它将在100NS范围内。



      以下为原文

    I see. so with a 10MHz SPI clock, it would be in the 100nS range.

使它变得更大，因为它需要10比特来接收一个值。



      以下为原文

   


make it more 1uS, since it will need some 10bits to receive a value

我想他的意思是从传输结束。



      以下为原文

    I think he meant from the end of the transmission.
 

我有这种感觉，但巧合是好奇的（无论如何，安定时间不应该与SPI传输速率有关）



      以下为原文

    I had this feeling, still the coincidence was curious (and anyway settling time should not be related to SPI transmission rate)

嗨，当电阻增加时有一个不希望的延迟，但是当它减小时不存在。我把数字锅的两端连接到3.3V直流电源，并把雨刷连接到示波器上。当罐的电阻从最小值变化到最大值时，当电压从0伏上升到3.3伏时，可以看到大约40ms的延迟。但是，当电阻从最大值变化到最小值时，没有延迟，电压从3.3～0V下降。代码很简单，并且在4MHz的SPI上每5ms发送一个值从0到127。这个值可以在2ms内完全改变。所以我预计延迟大约5ms。这是当电阻减小的时候，但不是在增加的时候。有什么想法吗？感激之情



      以下为原文

    hi,
 
There is an unwanted delay when the resistance increases, but not when it decreases. I have the two ends of the digital pot connected to a 3.3v DC power supply, and the wiper connected to an oscilloscope. When the resistance of the pot changes from minimum to maximum I can see a delay of about 40ms as the voltage rises from 0 to 3.3 volts. However, I do not get the delay when the resistance changes from maximum to minimum and the voltage drops from 3.3 - 0v.
 
The code is simple and sends a value from 0 - 127 to the pot every 5mS over SPI running at 4MHz. The value is capable of full scale change in 2mS. So I am expecting a delay of about 5mS. Which is the case when the resistance is decreasing, but not when it is increasing.
 
Any ideas please?
 
thanks
Tim

它可能取决于内部的一些高阻抗电路，例如，MOS栅控制电荷泵…你能从雨刷上张出一个波形吗？



      以下为原文

    It may depend on some high impedance circuits inside, say, MOS gate controlling charge pumps... Can you post a waveform from the wiper?

是的，我会。时间设定为50ms，图中显示上升时间和下降时间。跌倒后的颠簸是我意外地控制着。谢谢。



      以下为原文

    yes I can. The time period is set at 50mS. The picture shows the rise and fall time. The bump after the fall time was me moving the control accidentally.
 
thanks
   Attached Image(s)

数字罐在步骤之前有足够的时间吗？



      以下为原文

    Was the digital pot powered enough time before the step?

对。可以持续数小时，那就是上升和下降的波形。



      以下为原文

    yes. Can be on for hours, that that is the waveform that occurs for rise and fall.

嗨，我已经整理了一些我的延迟问题。但仍有2.5毫秒的上升时间延迟，附图显示上升时间延迟，并且很小的下降延迟。



      以下为原文

    Hi, I have sorted some of my delay problems. But there is still a 2.5mS rise time delay.
 
The attached picture shows the rise time delay, and very little fall delay.
 
thanks
Tim
 
   Attached Image(s)

看起来上升时间大约90% fs是非常短的，而最后10%的延迟占大部分延迟，这似乎更像1毫秒（如果范围是2.5毫秒/ div）。如果高电平是Vdd，则可能是数字电位器中的MOSFET在高侧没有得到足够的驱动，从而增加其电阻。看看10%和90%设置之间的转换是否更好。



      以下为原文

    It looks like the rise time to about 90% FS is very short, and the last 10% accounts for most of the delay, which seems more like 1 mSec (if the scope is 2.5 mSec/div). If the high level is Vdd, it may be that the MOSFETs in the digital pot are not getting sufficient drive on the high side, and thus increasing their resistance. See if the transitions between 10% and 90% settings are better.