FREE DIY open source BOOST driver!!! Tested & working!!

rhd said:

Is it? Step me through the math on that. I'm sure you're right.

I think it's a huge ripple pictured, but not really a startup spike, because isn't that just during normal operation? He said that those last three were the same shots, but "after" startup, which I would take not necessarily to mean a startup spike, but rather just the ongoing ripple. (which is just as bad)

Click to expand...

yeah i think it's an awful ripple too.



here it's 4 divisions from start to end of the signal so.

2.5us = 0.0000025 seconds.

F = 1/s

F = 1/0.0000025 = 400000 * 4 divisions = 1600000Hz = 1.6MHz

:wave:

foulmist said:

yeah i think it's an awful ripple too.

http://farm8.staticflickr.com/7226/7354385274_c9156a0448_z.jpg

here it's 4 divisions from start to end of the signal so.

2.5us = 0.0000025 seconds.

F = 1/s

F = 1/0.0000025 = 400000 * 4 divisions = 1600000Hz = 1.6MHz

:wave:

Click to expand...

Shouldn't it be divide by 4 divisions, not multiply by them?

If it's taking 4 divisions for the signal to repeat, then you shouldn't be multiplying by 4 (which gives a higher frequency) but dividing by 4 (which gives a lower one) ? No ?

I am pretty sure it's multiplied here is a tutorial on how to measure frequency of one full cycle

Oscilloscope Frequency Measurements

:drool::san::wave::toilet:
http://www.wisc-online.com/objects/ViewObject.aspx?ID=ACE3503

foulmist said:

I am pretty sure it's multiplied here is a tutorial on how to measure frequency of one full cycle

Oscilloscope Frequency Measurements

:drool::san::wave::toilet:
http://www.wisc-online.com/objects/ViewObject.aspx?ID=ACE3503

Click to expand...

Yep, I'm right

It IS multiplication, if you do it BEFORE the "Frequency = 1 / ...." calculation. But you reorganized your equation such that the multiplication came AFTER you had already divided. Order of operations. Multiplication becomes division if you move it to after a previous division.

So, following along with your tutorial (step by step, using their process):

One cycle is 4 divisions.
One division is 2.5 us, which is 0.0000025 seconds

Multiply: 4 * 0.0000025 s = 0.00001 s

Frequency = 1 / 0.00001 s
Frequency = 100 kHz


** I would point out that it's actually about 3.5 divisions, not 4 divisions, which changes the resulting frequency to more like 115 kHz. **

ah damn,

I am totally not thinking right now

anyway if it's 100kHz that means what? :thinking:

everything is worse than imagined?

Well, I would be thinking that maybe the wrong IC was used?

Well it has to be either 525kHz or 1.6MHz.

Tom, are you scoping the original design (the ben and rhd layout)?

Are there any shorts? Different parts? More details would be helpful.

I don't know how you guys are dong it...

But the way I've been taught how to measure frequency is to take the number of divisions from the peak of a wave, to the next wave, and then multiply it by the time base. Then take the inverse of that number to see what the frequency is.

Though, what I find odd is that the ripple is a sine wave.

It shouldn't be a sine wave... Sine wave means ringing. The ripple should be slopes, not sines. Think of how a boost driver works, there shouldn't be ANY ringing anywhere in there, and yet, this is what we're seeing right now.

Something is up here. I think either 1. Tom may not be scoping the correct area, 2. The oscilloscope might not be grounded correctly, or 3. The driver itself is messed up.

More scoping is needed, I think. If someone (rhd) like to send me a driver, I'd be more than happy to scope it with my analog one to see what the ripple looks like.

Le Quack:
I believe that is what I was doing, no?

Give it a shot. What do you end up with for frequency?

The following is incorrect because the divisions are 2.5 us, not the whole picture...

There are about 3 full waves in that picture. The full picture is 2.5 us. Therefore, each wave takes 2.5 / 3 = 0.83333 us to complete. That is, each wave has a period of 0.00000083333 s. In other words, there are 0.0000008333 s per wave. That means there are 1 / 0.0000008333 s/wave = 1200048 waves per second. The frequency is 1.2 MHz.

Click to expand...

I hope that helps. Thank you for designing this affordable driver. Hopefully this ripple will turn out to be nothing. If not, hopefully the fix will be easy. Cheers :beer:

I got the same reading as Scotoma, around 1.2 mhz.

I still believe there is something amiss, as I said, the ripple shouldn't be sine waves. Should be SLOPES, like a sawtooth wave of sorts. What we're seeing here is high frequency ringing.

Hey - I am pretty sure it's 2.5us per division, not for the whole plot.

Oh, I see. I was wrong then, the frequency is 1 / (3.5*.0000025) = 114285, or 114 kHz as previously stated. Sorry for adding confusion to the thread.

Wolfman29 said:

Hey - I am pretty sure it's 2.5us per division, not for the whole plot.

Click to expand...

Yep, it is. At least that is my understanding. And doesn't that end up giving you a frequency of around 115 khz?

Well, I will do this over more carefully when I get back and I will show a video of each step. I am brand spanking new to scopes so it is very possible I am making a mistake.

I did hook my PSU to the scope and adjust the voltage to make sure I was seeing what I expected to see. I was able to see straight lines at 5v, 8v etc...

Not much of a calibration or test, but at least I was seeing what I expected to see.

The last three photos, I am pretty sure they are just a shot of the running driver with various different scales.

The first ones are of the initial startup, and I want to do those again with a much longer timeframe.

LeQuack mentioned grounding. Is there anything special I should be doing? It is a three pronged cord and I'm assuming that it is grounded properly.

Also, I am hooking the probe up to the right side of the 1ohm resistor on Jufran's test load and the negative alligator clip that comes off the probe to the left side of the resistor.

Is that correct? I'd hate to be sending anyone on a wild goose chase because of my inexperience with a scope, but I'm trying to be very open about the fact that I have no idea what I am doing with the scope!

I'll see if I can upload the video and any other pics that I may have.

Here is the video, I will also look for any pics I have and add them to this post. I'll try to find one showing the chip.



I'm not finding any pics of the IC. I'll have to do those when I get home tomorrow or Monday. You have to think Moh ordered the right ic's. It is far more likely that I am doing something wrong with the scope.

I have three other scope pics. I'm not dead sure if these are of this driver, or the 537mA driver that I was running 8v into. I'll post them just in case...


Untitled by tsteele93, on Flickr



Untitled by tsteele93, on Flickr



Untitled by tsteele93, on Flickr