Accuracy of ZXLD1370 LED driver in buckboost topology?

[Flyback]

Hello,
We are very interested in the ZXLD1370 LED Driver IC, -in relation to work I am doing on LED beacons.
We like this product because it appears to offer accurate current regulation in spite of its relatively easier hysteretic control nature.
On page 16 of the ZXLD1370 datasheet, Equation 4 gives an mathematical expression for the LED current.
For this expression to be correct, (1-D) would have to equal GI_ADJ.
As you know, GI_ADJ is a constant set by a resistive divider, and (1-D) is variable, since D varies as either the input voltage or output voltage changes.
Therefore, how can equation 4 possibly be correct? As such, I cannot see how the LED current could be well regulated over load and input voltage changes. –Please could you offer explanation?
Please could you assure us that the LED current regulation will be good over possible input voltage changes from 11V to 30V, and possible output voltages anywhere between 23V and 33V? Our LED load is 9 LEDs with 1.5A in them, and the flash duty is 0.4, with flash frequency of 2Hz.

**broken link removed**

 

[ronsimpson]

Hi FlyBack,

No time now.
There are graphs of supply voltage and LED current that show good results. Not 0.01% but good enough for you eye. 5% or 2.5% like that.

There is a function inside that sets "the ratio of switch current to LED current". If the resistors are right you set this ratio and all is fine.
The formula does not take into account the ("error amp" and "adder") and and ratio setting.

 

[Flyback]

The ZXLD1370 claims to be able to regulate the led current to (as you say) around 5%, without directly sensing the led current, and this, to me , is surely impossible...when considering line and load regulation. The graph they show as figure 18, page 13 only shows vin between 7V and 17V........but what about up to 24V, say, does it go wrong above 17V?

Accuratly regulating LED current without actually sensing it directly is surely impossible?....if the ZXLD1370 really can do this, then its the only IC in the world that does it. Do we believe it?

 

[ronsimpson]

if the ZXLD1370 really can do this, then its the only IC in the world that does it. Do we believe it?

There are many ICs that regulate LED current some thing like this. This is seen in Diodes Inc parts.

The Inductor current can be measured within the accuracy of the current sense resistor and Vref. You can see that.
The LED current is the inductor current and the "D" or "1-D" depending on how you see the world.
....If D=50% then LED current is (inductor current X 50%)

Because we know the inductor current,
AND
we know the relationship between current in inductor and current in LED, I LED=I inductor X (1-D)
THEN
with a simple voltage multiplier we know the average current.

Another way to look at it:
If we know that D=50% then we know the inductor current needs to be 2x what we want in the LED.
It only takes a simple analog multiplier to get the results. Its in the IC.

 

[Flyback]

You are absolutely correct..but that was not the question.
The Duty cycle for a buckboost (whether diode or fet duty cycle) changes with variation in either vin or vout.
The ZXLD1370 purports to be able to keep led current the same in spite of this.
It claims to able to do this by use of the simple potential divider connected to GI_ADJ.
Surely this is not possible?

 

[crutschow]

.......................
Accuratly regulating LED current without actually sensing it directly is surely impossible?....if the ZXLD1370 really can do this, then its the only IC in the world that does it. Do we believe it?

It uses Rs to sense the current through the inductor which is proportional to the current through the LEDs. That's sufficient to regulate the LED current. Does that not make sense?

 

[Flyback]

Not to me because if either vin or vout changes, then D will need to change in order to keep the led current to the regulated value...the ZXLD1370 cannot possibly know what value of D is required for any given combination of vin and vout. The ZXLD1370 does not even measure vout, and even though its connected to vin, its not clear that it measures it.

As you said before, the LED current is (inductor current average)*(1-D).....and that is not the same as the inductor current.

 

[tvtech]

That's a pretty cool LED Driver. Brand new in the Market place. And price is reasonable.

And DigiKey have thousands in stock: **broken link removed**

Win win situation for all???
Maybe request a sample and test it. The Data/Specification file is pretty thorough....
Nothing better than actual testing though of the physical thing.

Regards,
tvtech

 

[Flyback]

Yes, but I think you are putting it mildly.....if the ZXLD1370 can keep LED current constant in buckboost, with either vin changing between 11V and 30V, or Vout changing from 23V to 33V, then the ZXLD1370 is quite simply a revolution in LED driving. There is n0 other chip that can do that and keep LED current regulation tight.. if the regulation over that vin and vout range is even as good as 5% then thats's still phenominal.

Its a shame the graph (figure 18) in the datasheet on page 13 does not take vin up to the upper automobile range of 30V (considering trucks). If it did then that would answer all questions. The fact that it does not go up to 30V, but stops instead at 17V, suggests to me that it maybe goes a bit pear-shaped above 17V..but I do not know. I need the eval board but none are in stock..
**broken link removed**

...If this chip can keep led current regulation tight over vin = 11v to 30V and vout over 23v to 33v,then we are talking revolution!....no other IC can do this in hysteretic mode.

 

[tvtech]

OK. I am out of my depth here. One thing I can tell you though is that as far as I can remember....anyone trying to do stuff that is compatible with both 12VDC and 24VDC systems always runs into trouble. It's either one or the other.

Regards,
tvtech

 

[alec_t]

the ZXLD1370 cannot possibly know what value of D is required for any given combination of vin and vout.

D is whatever is necessary in order to stabilise the average inductor current. That current is monitored and hence so is the LED current, indirectly.

 

[Flyback]

LM3421based LED driver can do it from 12v to 24v, but it has more in-depth feedback loop dynamics, and since the lm3421 is not hysteretic mode, it will be more difficult to avoid LED current overshoot when the flasher flashes back on twice per second. Also, PWM dimming with the ZXLD1370 in buckboost will have better resolution than with lm3421, since zxld1370 is hysteretic.
Question is, how accurate is the led current regulation in ZXLD1370.

"D is whatever is necessary in order to stabilise the average inductor current. That current is monitored and hence so is the LED current, indirectly."

...Sorry I don't know what you mean by "stabilise" the average inductor current?

 

[Flyback]

This is the response I just had from diodes.com

Thank you for contacting Diodes.
I have been passed your inquiry from Inquiries Europe.
Equation 4 shows that the LED current is independent of
input voltage and output voltage.
Equation 3 shows the Sense Resistor current IRS.
ILED is IRS*(1-D) so the (1-D) term in IRS is cancelled out
leaving ILED independent of D hence VIN and VOUT.
If VADJ=VREF then Equation 4 simplifies to
ILED=(VSENSE/RS)*GI_ADJ.

...they are just basically saying what it says in the datasheet, I am hoping someone here will know whats going in behind this?

I am sure you will appreciate that ILED cannot possibly be independent of D....as D changes, then if nothing else changes, then ILED will change.

 

[alec_t]

I don't know what you mean by "stabilise"

I mean 'regulate', 'keep constant', ......

 

[ronsimpson]

Look at the ZXLD1366. I can see how it regulates current. I have used parts like this.