Pulse Delay and Pulse Width??

[Digger28]

In understandable terms can someone describe the difference between these terms. I see that some PI machines have adjustable "width" and others adjustable "pulse". What does this mean and which is preferred?

Many thanks!!

-Rick

 

[Reg]

Hi Rick,

I am not sure just what your question is so I will try to answer what I feel you may be asking.

First, I will discuss the term delay and how it applies to a PI. A PI works on the basis of transmitting a strong pulse signal into the coil, shutting that pulse off quickly, and then waiting for a finite amount of time for the main signal to decay. This waiting period is called the delay. Once the pulse decay is near zero volts, a brief sample of the coil signal is taken and analyzed. If this sample is taken too soon, the amplifiers will be saturated and the detector will appear to not work, so there is a minimum delay which will work. Most detectors are designed to work within safe limits so the owner doesn't experience this failing condition.

This pulse/sample process is repeated at the rate of the repeating pulse signals commonly referred to as the pps or pulses per second. Now, this pulse into the coil can be short or long, depending upon the primary purpose. High conductive objects will or may generate or create a better response when a long pulse is used. Low conductive objects do not show the same increase, so a shorter pulse length can be used just as effectively when hunting low conductive objects such as small gold nuggets or gold rings, chains, etc. Why use a shorter pulse if both will work for gold? Well, battery life can be lengthened considerably by reducing the pulse length. So, there are advantages in deciding what might be the best pulse length for the application. Also, lengthening the pulse may not display that much of an increase, so the trade offs are not necessarily a simple decision.

Now, the delay is really this; it is the time between when the pulse is shut off and when a sample of the signal is taken. The shorter the delay, the stronger all signals will be. Also, low conductive objects such as small gold may only generate a brief response and that signal may decay before the sample is taken if there is too long of a wait. If this happens, then the object will not be detected. So, reducing the delay enhances the possibility of detecting smaller low conductive objects such as small gold. Once again, there is a minimum delay at which the detector will work correctly.

The pulse length is the time the strong pulse current is on. Generally speaking, a longer pulse allows the operator to better detect certain higher conductive objects. A shorter pulse length can is is quite common when trying to detect smaller gold or other low conductive objects.

I am not sure I have answered your question or not, but I have tried.

Reg

 

[Digger28]

Thanks Reg for the detailed explanation - that's great. So in summary the delay is the "dead" time between the end of the pulse and when the sample of the residual current is taken. The width is how long the actual pulse lasts. I appreciate that you embellished a bit with how these timing affect the operation of the machine.

-Rick

 

[Digger28]

Okay here's another dumb question. If the pulse rate is fixed at 600 pps for example it seems there would be a fixed allotment of time for each pulse cycle. If that's the case I'm wondering how one would adjust one of these attributes without affecting the other. In other words if I increased the pulse width would I not in effect be reducing the pulse delay period?

I think I'm missing something fundamental in my understanding. Help!

Thanks folks,

-Rick

 

[Reg]

Hi Rick,

You have to do the math to figure out what is what. Lets take a very simple example where the pulse rate is 1000 pps. This means there is 1 pulse each millisecond and this is the same as 1 pulse every 1000 usec. Now, if the pulse length is 100 usec, then the remaining time is 900 usec. If we wait even 25 usec and then sample for 30 usec we have just used up another 55 usec, so subtract that from the 900 and we have 845 usec of basic wait time before the next pulse. Ok, lets say this is a well designed unit that requires a second sample to eliminate the earth's field effect. So, lets say this happens 250 usec after the pulse ends. Keep in mind the main sample has already occurred so that time is included.

Now, since the pulse is 100 usec long and the sample is 30 usec, we are now at 250 +100+30 or 380 usec. This now means we still have 1000 -380 or 620 usec of wait time between pulses and samples. So, the detector is just waiting for this extra time to end before there is another pulse.

Now, since we have a lot of extra time between pulses, we can stretch the pulse length if we desire and still not change the pulse rate. We also have plenty of time so we can vary the pulse rate some if we need to. This may be needed to minimize external noises that may be almost the same or beating with the pulse sampling.

Now, also keep in mind that my selection of a 1000 usec pulse rate is a 1ms rate or stated another way, 1000pps. This is more pulses per second than the 600 you mentioned, so if a 600 pps rate was used, there would be even more wait time where nothing important was happening.

This all may sound confusing, but it does work out quite well. So, one can increase the delay or even the pulse length within reasonable limits and not have any impact on the pulse rate at all.

Reg

 

[parrott]

Reg... thank you for all the posts you have made regarding PI-types... Parrott

 

[Digger28]

Great write-up Reg, that makes perfect sense. I appreciate you taking the time to clarify this. It's a terrific plus to have knowledgeable folks such as yourself willing to share with the users.

Thanks !

-Rick