The PLD-LOM is now available as a bench-top precision instrument - having grown in specifications dramatically from the original starting point. I'm leaving the start of this thread intact because I think it makes for fantastic historical reading, and reminds me of how far things have come.

The current PLD-LOM7 is now available at NQRC - http://nqrc.com?vp=PLD-LOM7 ($479 AUD)

-----------------------------------------------------------------

Development type: For commercial production
Anticipated price: $149 AUD ~ $199 AUD
[FONT="Arial Black"]Current status: Prototype #3 in testing


Hello all,

This thread is for watching and discussing the development of the new PLDaniels LOM (Low Ohms Meter).

Please note, a LOM is not the same as an IRM (Internal Resistance Meter), although both do measure very low resistance values.

A LOM is like a normal Ohms meter / multimeter except that it's specialised for measuring resistances well below one (1) ohm.

What is a LOM useful for...

* Checking motor windings
* Checking for bad connectors
* Measuring actual resistance along power leads to determine losses
* Locating short-circuit faults on electronics

If you're serious about maximising the power out of your electric flight systems, then a LOM is very useful, it lets you sort out the good from the bad when it comes to wiring looms and connectors.

OK Paul, since you started this thread - I want to be officially put down as the first customer. Let me know what the final price is.

You know me - Gadgets-R-Us

:-)

John

John,

Happy to oblige. The easy part is over... now the hard, grueling task of picking a suitable casing as always looms over.

I could always do another acrylic sandwich, it'd be easy to make but the scuffing of the casing would make it look ugly really fast. Not really "toolbox" tolerant. . . . but it'd look good

Paul, given it's a more expensive product, how about this one with a laser cutout for the LCD?

It's Jaycar HB5610. Looks about the right size.

Handheld Electronic Enclosures - 135 x 70 x 24mm - Black

- Moulded in high impact ABS plastic- Available in light gray or black colour- UL-94 VO flame retardant material is available on request- Built in battery case will accept either 1x 216 9V type or 2 x AA (9V hardware supplied)- Transparent front panel available on request for IR remote applicationsWidth: 135Depth: 70Height: 24Material: ABSColour: Black

Originally Posted by pldaniels

John,

Happy to oblige. The easy part is over... now the hard, grueling task of picking a suitable casing as always looms over.

I could always do another acrylic sandwich, it'd be easy to make but the scuffing of the casing would make it look ugly really fast. Not really "toolbox" tolerant. . . . but it'd look good

John,

I've got a couple of those here, I was trying to test-fit the board/LCD into it but started to run into issues when trying to work out how/where to cram in the 4 banana sockets.

I'm hoping that with the revised PCB that I've got now, I'll be able to get the 4 sockets in.

Getting the LCD panel cutout is a possible but I'll still need to get a proper adhesive faceplate with a clear window for viewing.

What I really had hoped for (but most likely won't happen) is to laser cut out an entire face insert like those iR filters to fit the sockets and the LCD window.

An additional technical development too, I can drop in a 24-bit ADC in place of the current 16-bit one that I use, obivously quite a bit more expensive and realistically it won't give me more than perhaps 10~20x the resolution, despite the much higher bitcount (ultimately I'm limited by 1 component on the board which is a reference resistor).

Paul.

Hmm. So if I understand it the 24 bit DAC would give you maybe 0.1 to 0.05mohm resolution instead of 1mohm but that the absolute accuracy is limited by the reference resistor. Or is it a resistor noise issue? If it is only the absolute accuracy that is the problem, then I think the additional expense would be worth it. Often comparing two values one after another is just as important as knowing their absolute value. I imagine this might be useful to DIY motor folks (not something I know much about). Similarly for checking resistance drops in wiring harnesses. It's the abnormally high values you want to know about, not the absolute resistance to high accuracy. Just a thought.

John

Originally Posted by pldaniels

An additional technical development too, I can drop in a 24-bit ADC in place of the current 16-bit one that I use, obivously quite a bit more expensive and realistically it won't give me more than perhaps 10~20x the resolution, despite the much higher bitcount (ultimately I'm limited by 1 component on the board which is a reference resistor).

Paul.

Another thought. Can you apply the printed adhesive faceplate THEN send the case off for laser cutting the hole for the LCD. Avoids any "clear window" requirement.

Originally Posted by pldaniels

Getting the LCD panel cutout is a possible but I'll still need to get a proper adhesive faceplate with a clear window for viewing.
Paul.

Originally Posted by jj604

Hmm. So if I understand it the 24 bit DAC would give you maybe 0.1 to 0.05mohm resolution instead of 1mohm but that the absolute accuracy is limited by the reference resistor. Or is it a resistor noise issue?

The limitation is that I can get 0.1% resolution resistors at 10R (my reference resistor) but 0.01% ones don't start until 50R, however those 50R 0.01% resistors cost about $30 AUD each

What I may yet consider doing is providing space on the board for a 10~20 turn cermet trimmer, allowing me to use a normal 11R resistor or so and putting 10K trimmer in parallel which will then let me trim the 11R down to 10R. The downside with the trimming method is that trimmers, even good ones suffer annoying temperature coefficients and are twitchy over time.

Ultimately, the 10R 0.1% reference resistor alone is likely to be the best solution because it's moderately priced but also means that the accuracy of the device is still acceptable but shouldn't drift too much so that our precision will be acceptable too.

Paul.

I should point out that the actual value of the resistor doesn't matter much, rather -knowing- its real value and its temperature stability is the primary factor.

In the firmware I just type in the value of the resistor (in milliohms) into the source-code and the math does the rest.

Paul.

I managed to pull another great trick out of the hat and now I can simplify the circuit a bit more -and- offer a stepped, greater range with very little additional complexity.

Previously the LOM was going to be limited to about 16 or 32 ohms maximum (which is realistically quite a lot still, if you've got a wire showing 32 ohms it's quite a length!). Anyhow, now I'll be doing 0~16 ohms and then beyond 16 I'll skip up to the next range and go 16~100 (roughly), all without having to change the anticipated cost of the final product.

Paul.

Proto #2 has been submitted, I've made the design of this one such that I should be able to mount the 4mm banana plugs directly on to the PCB, that'll make things faster/easier/more-reliable on the assembly front.

The board is now slightly over 50x50mm so that it can mount directly onto the screwpoints in the case that I'm "planning" on using.

I have a suspicion that I'll need revision #3 but I'll be willing to be surprised

Paul.

Okay, no surprises... found two minor things on the PCB this morning :curse:

1. Didn't put a current limiting resistor on the power line to the buzzer (the buzzer is an option that'll let you do tracking of short circuits via audio feedback). Fortunately I can side-step that issue with ease.

2. I left 3 little solder-mask pads on the board as a carryover from an older revision by accident, fortunately they're not in critical places.


For the test leads, what I'll likely do is just stock pre-wired banana->alligator clip leads and banana->hook clip leads, then people can pick and choose what sort they want to use.

To do the short-locator audio test it'll be a case of needing -another- lead set, though a normal multimeter probe set should be fine in conjunction with the alligator clips. The test method will be that you attach the alligator clips to the two points where the short appears to be (eg, + and - rails on the faulty board) and then you can proceed to run the multimeter pointed probs along the track listening to the pitch of the audio feedback.

Paul.

LOM casing

Well, I finally managed to find a reasonably priced case for the LOM (and a few other similar projects no doubt). It has space in it for a 5x5cm board, directly mounted, accepts a 9v battery and fits into the hand quite nicely. The case also has a recessed front-face making it very suitable for a lexan adhesive panel.

The LOM PCBs (proto #2) are still "processing", at this point I don't expect to see them until the end of December - even though the proto mfg'er does seem to get them finished in ~5 days the delivery service is woeful.

Paul.

Here's a couple of pics of the LOM #1 resting in the casing I've selected for LOM #2. The PCB for LOM #2 is sized/drilled such that it should mount directly onto the 4 internal screws and also will have the banana sockets directly on the PCB as well.

I will have to get a lexan adhesive panel for this setup, it's an expensive option but there's no other way to finish it off as neatly.

Hi Paul,
I am enjoying this thread, thanks
I was wondering about creating a cheap, precision low resistance by using a length of calibrated resistance wire?
I fell over this Web page http://www.allaboutcircuits.com/vol_1/chpt_8/9.html
It made a new point for me "Kelvin Clips", they would certainly elimiante one of my doubts on low R measurement.

There is also a beautiful photo of a German standard 0.1 ohm resistor, alas posatge would be a killer!

Have a great day
Mark