Daqarta for DOS Contents



Free Registration

Contact Us
Daqarta for DOS
Data AcQuisition And Real-Time Analysis
Shareware for Legacy Systems
(Use Daqarta for Windows with modern systems)

From the Daqarta for DOS Help system:


If the trigger Source is Intern in Board mode, or Main, AM, or FM in Virtual Source mode, then Slope works together with trigger Level to specify the trigger point on either the rising (Pos) or falling (Neg) portion of the signal, or alternating between these (Alt). A vertical bar at the right edge of the trace area indicates both level and slope. If the slope is Pos or Alt, the bar starts from the bottom and extends up to the trigger level. If the slope is Neg, it starts at the top and goes down.

Alt would not normally be used where you want a stable waveform display, but occasionally can be helpful to spot signal asymmetries. You can use persistence of vision to view both phases superimposed, or you can use averaging to see if they average to zero.

If Source is Extern (Board mode), then Slope still behaves as above but now refers to a digital signal edge... a very steep slope.

With Source set to Stim, Slope vanishes unless Sequential mode is active and the Stimulus Generator is inactive, in which case it refers to the polarity of the digital trigger Pulse output. Here Alt can be used with averaging to isolate the portions of a response that are unchanged by stimulus polarity from those that invert, since the latter will average away to nothing.


When internal level triggering is active, a trigger occurs whenever the input signal reaches or crosses the threshold set by Level while going in the direction specified by Slope. The level control is not available if Source is Extern or Stim.

A vertical bar at the right edge of the trace area gives a quick visual indication of the trigger level. If Slope is Pos or Alt, the bar comes up to the trigger level from below, or if the slope is Neg it hangs down to the level from above. The bar reflects the trigger level relative to full scale only, and does not react to changes in trace magnification.

For Auto mode, the actual trigger level is adjusted after each sweep so that it is in the same proportion to the signal peak as Level is to full scale. The bar remains at the Level value and does not track the signal peak.

The default trigger level of zero is a general purpose value, since most input signals pass through zero. But you will often get better results by setting a larger value, positive or negative, depending upon your signal. If each waveform cycle has multiple zero crossings, a level set at zero could trigger on any one of them. Similarly, if the input spends a lot of time near zero, setting the trigger level there will result in spurious triggers from noise which goes back and forth through zero.

For example, suppose you are looking at the response of a system to a recurring impulse... say a hammer striking a bell, or a plucked guitar string. The response will have large oscillations that decay toward zero. You would set the trigger level just below the highest peak in order to avoid triggering on lesser peaks or random background noise between impulses.


Sometimes the region of interest in a signal is not near the trigger point in time. With an impulsive response, you probably will want to set Slope and Level to trigger near the initial peak of the response for stable triggering. You would normally expect to see the next N samples after that. But if you want to study low-level phenomena that take place as the vibrations die down, you may have to wait for many samples. You don't want to slow the sample rate enough so that the entire response duration is visible all at once, or you may compromise time resolution and introduce aliasing.

The ideal way to handle this is to slide the trace area later in time, to look at events that happen well after the trigger while still maintaining a high sample rate. This is done by setting a positive trigger Delay value. After the trigger is detected, Daqarta will wait for the equivalent number of Delay samples before starting to collect the N samples for the trace, thus sliding the "viewing window" to later times.

The waveform display X-axis will always show the correct time since the trigger sample. With a spectrum display, there is no change to the frequency axis since Daqarta is analyzing the same number of samples at the same sample rate... they just start at a different time.


But what if you want to see events that happened BEFORE the trigger sample? If you set the trigger level to a high value to get reliable triggering from a response peak, you will miss seeing the leading edge of that peak. Or consider the case of a normal electrocardiogram (EKG), where there are low level P and Q waves that precede the large R wave trigger event.

CAUTION: Do not connect any electrical equipment to a living subject without proper signal isolation techniques. A lethal shock could result.

You can look ahead in time with a negative Delay value. This is not really precognition, since you don't get to actually see these events before they happen... you get to see events that happened before the trigger happened. In reality, incoming data is simply stored to a large circulating buffer, and when the trigger arrives it is an easy matter for Daqarta to count back the proper number of samples in the buffer to get data before that trigger.


Daqarta can operate with delays of more than +/- 30000 samples. But note that the acquisition time includes all of the samples, even those you don't see. For example, at a 20 kHz sample rate a delay of 30000 samples adds 1.5 seconds to the total time between trace updates.

TRIGGER CYCLE (Board mode only):

With Intern or Extern trigger sources, Cycle sets the minimum time between valid triggers. This is analogous to "hold-off" in conventional oscilloscopes. To understand how this might be useful, consider a signal that is a repeating tone burst. If you want to trigger on the start of the burst, you can set the trigger Slope and Level to select a point on the rising edge of the first cycle of the tone.

But every other cycle of each burst will also have an identical point that could generate a trigger. If the tone burst is longer than the trace, the next trigger will come in the middle of the same burst. If the tone burst is short enough that more than one burst is shown in each trace, you could get a trigger in the middle of some other burst. Either way you could get a trace that jumps from wave to wave on successive sweeps.

With long tone bursts, set Cycle to a time that is longer than one burst but shorter than the time between burst starts. Then Daqarta will ignore all triggers until the dead time between bursts. Since there are no triggers there, it will wait for the start of the next burst. With short tone bursts, the procedure is similar except the Cycle value should be set slightly less than a multiple of the burst cycle time.

For example, suppose the bursts are 15 msec long with 5 msec between them, for an overall burst cycle time of 20 msec. If the sample rate is 20 kHz and N is 512, each trace will show 25.6 msec, ending 5.6 msec into the next burst. You would set Cycle to just under some multiple of 20 msec, like 39, 59, 79, etc. Daqarta will then start looking for triggers only in the dead time, and accept the first trigger of the next burst.

Setting this up usually requires some "fiddling", especially in Sequential mode. Here the processing time between traces is unavailable for trigger detection, so if you set Cycle too close to the sweep duration Daqarta can't respond at the time you intend... and by the time it can respond the signal might be into the next burst. If you set Cycle too low, the Speed indicator at the upper right above the trace will show SLOW. You can then keep increasing Cycle in multiples of the input cycle time until the Speed shows SPEC, meaning you are actually getting the cycle time you have specified.

Scrolling the Cycle value in Sequential mode may be unsatisfactory, since the key processing time will add to the normal processing time and bias the result... direct entry will probably work better.

In RTime mode, there is no limitation due to processing time, so you not only can pick values closer to the end of the trace, but you can scroll the Delay value as needed. The Speed display will show a GAP value, not SLOW or SPEC. The GAP value will be a constant at low Cycle times, until you advance Cycle beyond the sweep time. Then GAP will start to increase and and you will know that the Cycle time is in spec.

A particularly useful application of Cycle is in adjusting Overlap in Free-Run RTime mode. Normally there is a certain amount of trace-to-trace variation in processing time, which causes different OVL values on each trace. By using Cycle set at least as long as the slowest processing time, the OVL values can be made constant. This is easily done by eye, scrolling the Cycle value while watching OVL.

When Cycle is set equal to the trace time (the maximum value shown on the waveform display X-axis), the display will show 0 GAP. One step below that will give 1 OVL, and each additional step lower will increase OVL by 1, up until the point where Cycle is too short for the procesing time and the OVL value starts to jitter.

You can also try Cycle with signals such as FM, where there is no dead period to allow true synchronization. Here you would just set Cycle as closely as possible to an exact multiple of the input period. You won't get a stable trace, but you may be able to slow down the rolling enough to be useful.


If the trigger Source is set to Stim or Pulse, then Cycle is not available in RTime mode. In Sequential mode, it is used to control the stimulus repetition rate by delaying each sweep as desired. Its function is similar to the above "hold-off" trigger Cycle, but the value can be set separately.

The RTime repetition rate is controlled by the length of the stimulus cycle set via the Stimulus Generator control menu.


Questions? Comments? Contact us!

We respond to ALL inquiries, typically within 24 hrs.
25 Years of Innovative Instrumentation
© Copyright 1999 - 2006 by Interstellar Research