Analog Vs. TTL.....

[300EVIL]

Okay, the way I see it is analog lets you adjust power level of your laser all the way to off and TTL is basically on or off. Right? So, I have heard that you cannot feed an analog laser driver with TTL. How would I go about converting a TTL signal to analog. I know it's a step backwards but I don't have a choice as of yet and want to upgrade certain components for the future. Any suggestions?
Thanks in advance!
Adam

[sonaluma]

Hi Adam,

Most analog LASER power inputs expect 0V (full off) to 5V (full on) signal, and like you say, anywhere in between is OK. Most TTL LASER power inputs expect 0V (off) or 5V (on) signal.

So check with your manual (to confirm the correct voltage range), but it is likely there is no problem to connect the analog modulated LASER to your TTL output, it will act like TTL modulated LASER in this case.

Kind regards,

sonaluma

[dave]

You will have no issue driving a analogue laser with a ttl signal. You just wont get the full advantage of having a analogue laser, however should you upgrade you laser software in the future, a analogue laser will be a million times more useful.

[300EVIL]

Hi Adam,

Most analog LASER power inputs expect 0V (full off) to 5V (full on) signal, and like you say, anywhere in between is OK. Most TTL LASER power inputs expect 0V (off) or 5V (on) signal.

So check with your manual (to confirm the correct voltage range), but it is likely there is no problem to connect the analog modulated LASER to your TTL output, it will act like TTL modulated LASER in this case.

Kind regards,

sonaluma

Hi sonaluma,
Thanks for confirming that information. I pulled the datasheet on the Wavelength Electronics PLD-5000 LD driver I have and it says in bold.
"This input was designed for analog signals only and should not be used with TTL signals." I have also confirmed this is a 0 to +5V signal. Weird huh...
Thanks!
Adam

[sonaluma]

Hi Adam,

When a diode controller is designed for TTL modulation only, it may be possible for it to operate faster in this mode than it's analog counterpart. e.g. a typical TTL modulated laser may be able to blank the beam at about 10kHz, whilst a typical analog modulated laser may be able to reproduce a 10kHz sine wave, but it could not completely blank the beam at this frequency. Of course, this is a big simplification and there are many other factors which may affect the modulation frequency (e.g. the optical components)!

So, I think the manufacturers warning is just that your modulation frequency using this analog modulated laser in "TTL mode" may be limited (but it is probably still good enough for your light show). I don't think you will damage the system putting light show blanking frequencies on the analog input.

Kind regards,

sonaluma

[300EVIL]

Hi Adam,

When a diode controller is designed for TTL modulation only, it may be possible for it to operate faster in this mode than it's analog counterpart. e.g. a typical TTL modulated laser may be able to blank the beam at about 10kHz, whilst a typical analog modulated laser may be able to reproduce a 10kHz sine wave, but it could not completely blank the beam at this frequency. Of course, this is a big simplification and there are many other factors which may affect the modulation frequency (e.g. the optical components)!

So, I think the manufacturers warning is just that your modulation frequency using this analog modulated laser in "TTL mode" may be limited (but it is probably still good enough for your light show). I don't think you will damage the system putting light show blanking frequencies on the analog input.

Kind regards,

sonaluma

AH! Okay, now this all makes perfect sense! Thanks for clearing that up for me, I just wasn't getting it. So as long as I don't try to pump a couple MHz of modulation to it, it "should" be able to handle a a couple kHz or so for laser light show blanking.. Perfect! I was scratching my head wondering why such an expensive driver couldn't handle such a simple task.
Thanks Again!
Adam

[OPO226]

Hi sonaluma,
Thanks for confirming that information. I pulled the datasheet on the Wavelength Electronics PLD-5000 LD driver I have and it says in bold.
"This input was designed for analog signals only and should not be used with TTL signals." I have also confirmed this is a 0 to +5V signal. Weird huh...
Thanks!
Adam

Strange. Part of the problem may be in the meaning of "TTL." TTL was an older family of logic chips. The description "TTL" can be applied to an input or output of a device which behaves equivalently to TTL chips' input levels and input/output drive capabilities. TTL inputs consider logic low to be <= 0.8V. Logic high is =>2.0V. The inputs of true TTL chips source current! That means a current (typ. 0.5mA) flows out of the input when the input is pulled low. When pulled high, then input (I think) draws no current. If left open, the input pulls itself high because it sources current, which pulls itself toward positive voltage.

The outputs of TTL don't source current well but can sink current. The outputs can pull below 0.4V even when sinking a few mA, and up to 20mA can be driven into them, for LED driving for instance, without trouble. A high level output from a TTL chip is only about 3.6V when the chip is powered from 5V. It also will drop quickly to a lower voltage when just a small amount of current is drawn from it. This is why "pullup" resistors are frequently found in TTL circuits, which assist outputs to rise all the way to 5V when high.

All of this is somewhat obsolescent, as modern HC CMOS and similar CMOS families have close to "symmetric" output drive, meaning they can sink and source equal currents, and they swing almost completely to 0V or 5V when low or high respectively. CMOS inputs draw effectively no current, except for a tiny charging current since they have about 4pF of input capacitance. This only becomes relevant at high switching frequencies.

If an input is labelled as "TTL" then it should behave like a TTL chip input. Unfortunately, the term is often used very loosely to refer to a digital input or output, without regard to whether the true TTL IO characteristics are modeled.

Furthermore, understand that digital (squarewave) signals are just a type of analog signal, of the step function variety. They have a lot of high frequency spectral content due to the fast rise/fall times. Thus, for ex. if a HC-CMOS output drives an analog device with a 0-5V input voltage range with its typical 5ns rise/fall times, the analog device would need an input bandwidth of 70MHz to follow the edges. So what would happen if an analog device didn't have this much bandwidth? We would expect that a decently designed analog input circuit for whatever would not have trouble with this, but would simply low-pass filter the signal to whatever the device bandwidth is. Thus, a 30kHz analog modulation laser would be expected to show an optical output rise/fall time of 11.7us regardless of the speed of the input edges for any input rise/fall times faster than 11.7us. Correspondingly, it would be necessary to test the analog input of a modulated laser with rise/fall times much faster than 11.7us to truly measure the laser's inherent modulation bandwidth.

BTW, the formula for rise/fall time in relation to analog bandwidth is:

tr = tf = 0.35 * fc

where tr = tf is the rise/fall time in seconds
and fc is the cutoff frequency in Hz, ie., the -3dB bandwidth of the analog system. This is based on the standard definition of rise/fall time as being measured at the 10% and 90% levels of signal swing. Modern digital scopes can measure this automatically.

There are rare circumstances I suppose where an analog front end might be unstable when presented with input signals with spectral content higher than some recommended value. But it would be a poor design to allow such an input. If the internal circuitry had such a limitation, then it would be wise to design in a low-pass filter to restrict input spectral content to safe levels. Note that there needn't be an explicit input low pass filter for an analog system to have limited bandwidth. Most of the time the low-pass bandwidth limit is inherent in the design of the circuitry.

I would be quite surprized if the Wavelength driver couldn't deal with a "TTL" input signal. Oh, I suppose there is one other issue. A true TTL chip driving an analog thingy is kind of silly because the chip has such a non-linear output drive (which may present another obscure possibility of peculiar instabilities). I always buffer digital signals going to the outside world with a pair of HC CMOS gates in parallel with a 27 ohm resistor or so in series with the output connector. This forms almost a perfect 50ohm cable driver.

[300EVIL]

DAMN! Somebody has a masters degree in digital electronics don't they! Wow, thanks for the reply however, most of what you said is way over my head. My synopsis is that I may want to consider adding a low pass filter to the signal in order to keep the rise and fall of the TTL from making the sensitive analog reception from thinking otherwise.... Right?
Thanks!
Adam

[OPO226]

DAMN! Somebody has a masters degree in digital electronics don't they! Wow, thanks for the reply however, most of what you said is way over my head. My synopsis is that I may want to consider adding a low pass filter to the signal in order to keep the rise and fall of the TTL from making the sensitive analog reception from thinking otherwise.... Right?
Thanks!
Adam

You should be safe feeding the Lightwave with signals derived from digital sources with a low-pass filter. You could also put a 5.1V zener diode across the input as a protection from voltage swings outside the 0-5V range.

I gotta figure out how to have a sig....

Good day!

[300EVIL]

You should be safe feeding the Lightwave with signals derived from digital sources with a low-pass filter. You could also put a 5.1V zener diode across the input as a protection from voltage swings outside the 0-5V range.

I gotta figure out how to have a sig....

Good day!

Go to "User CP" at the top then go to "Edit Signature" when you get there. That I know how to do.
Thanks Again for the input. I will add a low pass filter and a Zener. It never "hertz" to be careful! (pun intended....)