AlanB's Drop In Dream Come True - Regulated Driver!!

[LuxLuthor]

I noted the med setting had an ongoing fluctuation range, but low and max remain fixed Vbulb.

Freshly charged Emoli cells, rested about an hour are in a range of 4.135 to 4.158, depending on the cell....so call that 8.27 to 8.28V They always drop to around 4.175V after just a few minutes off charger. I always use the Fluke 179 in DC mode.

 

[LuxLuthor]

This is almost enough to make me want to upgrade to a true RMS AC+DC DMM. What is the "cheapest" Fluke that can do that again? I forget which page of which thread has that discussion buried.

 

[Alan B]

This is almost enough to make me want to upgrade to a true RMS AC+DC DMM. What is the "cheapest" Fluke that can do that again? I forget which page of which thread has that discussion buried.

I'm not sure which is the "cheapest". The one I have is not low cost and it is discontinued but available. Any meter with an RMS mode on the DC range should be okay. RMS mode on the AC range is not what we need.

The calibration MUST be do-able with a standard meter, so we need to work through this, and using a regular averaging meter is a good way to check this. I have both types of meter, so I can do some verification as well.

The first step is to see if we can duplicate the 2.0 / 7.3 volt readings with an averaging meter and calculations. I recall the 2.0 reading coming out a little high, but the 7.3 was very close. The narrower pulse widths are less accurate due to the way the meters work. That is fine as we need accuracy at the high end rather than the low end. I was not clear which range/mode your meter was in during the above measurements. It should be in the DC (averaging) mode. You can measure battery voltage from one of the bipins to the socket mounting screw.

It will be more accurate if these measurements are made with the bulb out. I stick short insulated hook-up wires (with stripped ends) into the bipin sockets and attach them to the meter probe clips.

 

[Alan B]

I adjusted the voltage to agree with the 7.30V reading that I was getting when I was calibrating and using your average DC readings:

7.67V supply (instead of 8.0):

2.10V
7.30V

This agrees with my memory for the 2V setting as well, it was very close to 0.1V high.

If we can get these readings to work without the bulb we should be able to calibrate.

 

[milkyspit]

This is almost enough to make me want to upgrade to a true RMS AC+DC DMM. What is the "cheapest" Fluke that can do that again? I forget which page of which thread has that discussion buried.

Lux, I doubt it's the cheapest suitable Fluke, but my Fluke 189 does that as well as data logging and a number of other nice things. They pop up on ebay from time to time.

 

[wquiles]

Lux, I doubt it's the cheapest suitable Fluke, but my Fluke 189 does that as well as data logging and a number of other nice things. They pop up on ebay from time to time.

I agree. The cheapest model that I found that can measure the Vrms directly (using the DC+AC mode) is the Fluke 189, with the Fluke 289 also being capable, but costing more money. I have one of each (189 and 289), and I got both of them on Ebay for a "reasonable" price (the 189 was used, the 289 was new).

Will

 

[Alan B]

Looking at the extended specifications document for the Fluke 187/189, it appears that the 187 is also a candidate. They show the same specs for both, and they include AC+DC accuracy ratings up to 20khz.

 

[LuxLuthor]

It will be interesting and likely useful for others who will be using "regular" DMM for to now be able to check these readings with both Fluke 179 & 189. That should give people some "rule of thumb" guidance on what their DMM readings correlate to with 189.

 

[wquiles]

It will be interesting and likely useful for others who will be using "regular" DMM for to now be able to check these readings with both Fluke 179 & 189. That should give people some "rule of thumb" guidance on what their DMM readings correlate to with 189.

From what I have read, and from my own bench experience with various Fluke's with and without the AC+DC mode, you can't really correlate very well the Fluke 179 and the Fluke 189 when measuring the PWD signal due to the varying duty cycle. How well it would correlate would depend on how close to a 50% duty cycle the PhD regulator would be running during your measurements. The RMS function on a "normal" True RMS Meter applies to a duty cycle around 50%, which is the narrow range in which the 179 would measure reasonably close. I experienced this on my Fluke 87 almost 3 years ago when testing the PWD driver from Willie Hunt, and being frustrated about the lack of "accuracy". You can see in this post how I am using the RMS equations to turn the non AC-DC measurements into the approx. value of the true RMS voltage:
https://www.candlepowerforums.com/threads/125451&highlight=Fluke

Once you depart from the 50% duty cycle (towards 0 or towards 100), then only meters with the AC+DC will be able to measure accurately, although once you get in the duty cycle in the 0-5%, not even the 189 measured correctly: I seem to remember it was something to do with the crest factor exceeding 6, or something like that. I "captured" this problem with the 189 last year in Alan's original collaboration thread, including this older post, where the lowest level was not even close, the next level up was close, and then on it was accurately giving me the expected voltage:
https://www.candlepowerforums.com/posts/2650581&postcount=15

Will

 

[niner]

I connected Jimmy's JM-PhD-D1 on my bench, with a resistor (1K) as dummy load, powered by a bench PS (21V). This particular regulator is setup as 17.5V by Jimmy himself. I measured the output with a Fluke 187 and a Tektronix TDS3052 scope at the same time. Here are the pictures:

Scope report the output at 17.56V RMS

While my DMM measured at 15.399 V.

 

[wquiles]

Right now your Fluke is setup for plain vanilla DC voltage measurement, which can't possibly account for the PWM signal's duty cycle.


You need to set the Fluke 187 to the special DC mode in which the display will show "AC + DC". To get into that mode, press the small blue button on the right side a couple of times, until the display shows "AC+DC". Then the Fluke will display the proper value, which will be around 17.5V.

Will

 

[niner]

Thanks Will! I got this meter used, never really learn how to use it.

 

[Alan B]

Right now your Fluke is setup for plain vanilla DC voltage measurement, which can't possibly account for the PWM signal's duty cycle.


You need to set the Fluke 187 to the special DC mode in which the display will show "AC + DC". To get into that mode, press the small blue button on the right side a couple of times, until the display shows "AC+DC". Then the Fluke will display the proper value, which will be around 17.5V.

Will

Good catch, Will.

 

[wquiles]

Good catch, Will.

No problem. Just trying to help

Will

 

[LuxLuthor]

From what I have read, and from my own bench experience with various Fluke's with and without the AC+DC mode, you can't really correlate very well the Fluke 179 and the Fluke 189 when measuring the PWD signal due to the varying duty cycle. How well it would correlate would depend on how close to a 50% duty cycle the PhD regulator would be running during your measurements. The RMS function on a "normal" True RMS Meter applies to a duty cycle around 50%, which is the narrow range in which the 179 would measure reasonably close. I experienced this on my Fluke 87 almost 3 years ago when testing the PWD driver from Willie Hunt, and being frustrated about the lack of "accuracy". You can see in this post how I am using the RMS equations to turn the non AC-DC measurements into the approx. value of the true RMS voltage:
https://www.candlepowerforums.com/threads/125451&highlight=Fluke

Once you depart from the 50% duty cycle (towards 0 or towards 100), then only meters with the AC+DC will be able to measure accurately, although once you get in the duty cycle in the 0-5%, not even the 189 measured correctly: I seem to remember it was something to do with the crest factor exceeding 6, or something like that. I "captured" this problem with the 189 last year in Alan's original collaboration thread, including this older post, where the lowest level was not even close, the next level up was close, and then on it was accurately giving me the expected voltage:
https://www.candlepowerforums.com/posts/2650581&postcount=15

Will

Love all the great feedback. lovecpf

 

[KiwiMark]

How are the newly arrived regulators going - what setups are you using them with?