I just put FoulMisT's FMT driver on an oscilloscope...

[foulmist]

60mA ripple? WTF? That is a lot isn't it? You've got to be making an error somewhere...

I am not totally sure if those numbers are correct.
It should be lower than 60mA at 1.7A. Somewhere around 20mA max but I might be wrong.

at 500mA it would be close to 1mA to 5mA tops.

and about the power source. at close to 2A output the input current is from 3A to 3.5A that's to be expected. In boost mode input current is always greater than the output in order to boost the voltage.

the TPS63020 is 4A switch buck boost so it's totally normal.

the new revision features lots of improvements in that particular area. At 1.7 the current ripple should be close to 5mA tops. I am getting them made soon. Will send one to Tom to scope too.

 

[bobhaha]

I am not totally sure if those numbers are correct.
It should be lower than 60mA at 1.7A. Somewhere around 20mA max but I might be wrong.

at 500mA it would be close to 1mA to 5mA tops.

and about the power source. at close to 2A output the input current is from 3A to 3.5A that's to be expected. In boost mode input current is always greater than the output in order to boost the voltage.

the TPS63020 is 4A switch buck boost so it's totally normal.

the new revision features lots of improvements in that particular area. At 1.7 the current ripple should be close to 5mA tops. I am getting them made soon. Will send one to Tom to scope too.

That's good to hear! 60mA swing is a lot... I've been doing some work with the same converter and not getting anything like that with my simulations...

 

[lasersbee]

I may have missed something...

Where on the Driver exactly are you connected when taking those
Scope readings...:thinking:


Jerry

You can contact us at any time on our Website: J.BAUER Electronics

 

[tsteele93]

OK, I am no longer going to make any comments on the results of the tests as I am clearly not qualified. So far, I haven't seen anyone who really seems to understand these who has commented. I'm a little disappointed that there isn't an oscilloscope expert on here to help. Maybe Cyparagon and we aren't giving him good data to work with?

With that in mind, here is a bunch of shots to show what I am seeing...

First of all, my setup.

I have a Jufran test load set to 6 diodes.
I have the ground lead of the scope hooked to the left side of the 1 ohm resistor and the probe is hooked to the right side of the resistor.

Much like this picture of when I was testing the BenBoost...


Untitled by tsteele93, on Flickr

Other than that, not much to say. I do have the multimeter also attached to the left and right side of the resistor, but I have removed it in the past and not seen any difference in the waveforms. I assume it has such a high impedance that it won't make any difference. It shows the 1.69-1.74 voltage each time I fire up the driver.

Cyparagon mentioned showing the scope with no load. So here it is with the probe and ground hooked to the circuit with no power being applied.


Oscilloscope noise - connected to test load with no power. by tsteele93, on Flickr


Oscilloscope noise - connected to test load with no power by tsteele93, on Flickr

Now here are the results of the FMT driver...


FMT 1.75A driver 50mV/block 250ns by tsteele93, on Flickr

Here it is at a higher resolution on the voltage scale - 20mV/block


FMT 1.75A driver 20mV/block 250ns by tsteele93, on Flickr


FMT 1.75A driver 20mV/block 500ns by tsteele93, on Flickr

Next, here are some DC shots, at 1.75V to show different timebases and what they look like. Starting with 100ns...


FMT 1.75A driver 500mV/block 100ns DC Coupling by tsteele93, on Flickr


FMT 1.75A driver 500mV/block 50ns DC Coupling by tsteele93, on Flickr


FMT 1.75A driver 500mV/block 25ns DC Coupling by tsteele93, on Flickr



FMT 1.75A driver 500mV/block 10ns DC Coupling by tsteele93, on Flickr


FMT 1.75A driver 500mV/block 5ns - as low as my scope will go by tsteele93, on Flickr

 

[Cyparagon]

Just like you would with cameras, it's better to zoom in on the area of interest. You want the waveform to vertically fill the screen to provide more detail. Since the wave is repetitive, you should also expand the time-base to only fit a couple waves.

It's close enough I can say it's a decent measurement. The switching-frequency ripple isn't great, but can be greatly reduced with a low-ESR filter cap.

The much larger deviation in the second-to-last image is a little concerning. Not bad enough to kill a 445 diode, but bad enough you should look into it.

 

[foulmist]

It's already been taken care of in the next revision (that's why revisions are a good thing - to fix imperfections). And btw I have tested this with a LPC-826 red too at 480mA.

I have even rapidly switched it on and off by connecting the input wires by hand - touching the terminals of the battery and power supply creating a spike on where the wires touch and it's still working pretty good. Not led not killed.

I haven't got the chance to try this with a 12x Bluray as my luck with the blurays is simply - the worst in the world.

 

[tsteele93]

Just like you would with cameras, it's better to zoom in on the area of interest. You want the waveform to vertically fill the screen to provide more detail. Since the wave is repetitive, you should also expand the time-base to only fit a couple waves.

It's close enough I can say it's a decent measurement. The switching-frequency ripple isn't great, but can be greatly reduced with a low-ESR filter cap.

The much larger deviation in the second-to-last image is a little concerning. Not bad enough to kill a 445 diode, but bad enough you should look into it.

This is a 60Mhz scope, and when I get to 50mV (I think and higher resolutions) I get a BW to the right of the numbers. Do you know what that means? I'm assuming BW means band width but maybe not, and if so I'm not sure what it is saying about the band width.

Thanks!

 

[Cyparagon]

I think it's warning you that the numbers might not be exact at that sensitivity - especially considering the noise you're getting at that level.

 

[tsteele93]

That is what I was thinking... But I can't find ANYTHING in the manual on it. Isn't that odd?

Here is an online copy...

TDS 200 User Manual

Thanks for the input.

EDIT: here is something from the manual

VOLTS/ DIV
Range 2 mV/ div to 5 V/ div at input BNC
(Full bandwidth at > 5 mV/div to 5V/div, 20 MHz at 2mV/div to 5mV/div)

I wonder if it is saying the bandwidth is reduced to 20Mhz? I know it is happening at 50mV and 20mV but maybe using a 10x probe changes that?

 

[Cyparagon]

I was wrong. See page 101. "Limits the bandwidth to reduce display noise." So it acts as a low(ish) pass filter. Shouldn't effect your measurements for most cases, but since those peaks are very short, it's worth having a look. It can be turned off in the same menu where you change coupling.

 

[tsteele93]

Hmm, I thought that I had that turned OFF but I will double check tomorrow. Thanks again for helping sort this out. +1