A useful mod to the Revo "Honesty Meter" - RC Groups
Shop our Airplanes Products Drone Products Sales
Thread Tools
Jun 23, 2017, 11:31 PM
ancora imparo
jj604's Avatar
Discussion

A useful mod to the Revo "Honesty Meter"


Revolectrix have just introduced an IR meter that operates on the same principle as Wayne Giles ESR/IR meter but incorporates the multi cell display of the (unimpressive and now discontinued) Turnigy IR meter. The Revo meter doesn't give exactly the same values as the ESR/IR meter but it's a lot closer and more consistent than the Turnigy.

The Revo "Honesty meter" comes as two PC boards screwed together and just encased in thin transparent heat shrink. Cheap to make but hardly protects what should be a precision instrument.

I noticed that it uses the same LCD display as my Turnigy IR meter, and with some minor work the Revo meter can be put in the Turnigy case which offers it good protection if it is tumbling around in the field box.

You need to remove the 10A fuse, bend the terminals at 90˚ and solder it back. Also do some minor work to the end case cutout. The whole thing fits firmly in the case and won't move. I put a small plastic spacer at the end where the fuse is to make it all align perfectly.

John
Last edited by jj604; Jun 24, 2017 at 02:54 AM.
Sign up now
to remove ads between posts
Jun 24, 2017, 01:58 AM
Registered User
john4565's Avatar
That's a nice modification John; good job.
Jun 24, 2017, 01:25 PM
Unregistered User
vimy g eaou's Avatar
Just looked at the Revo IR meter.

You're right John, it does look rather sad in it's cheap and nasty cover. A trip to Jaycar for a decent project box will fix that if you don't have the Turnigy meter. Some buttons wouldn't go amiss either.

Interesting that Revo seems to be adopting the use of XTxx connectors. There is a small Revo MPA board that has them mounted now.
Last edited by vimy g eaou; Jun 24, 2017 at 01:32 PM.
Jun 24, 2017, 02:26 PM
Registered User
"The Revo meter doesn't give exactly the same values as the ESR/IR meter but it's a lot closer and more consistent than the Turnigy."


Well, exactly how far is it off?
Give a numerical example please
Are you measuring at 72 deg F, pack rested at least 2 hours?
Jun 25, 2017, 07:04 AM
ancora imparo
jj604's Avatar

Revo Honesty meter and Turnigy IR meter values


I normally test at 20℃ as my standard - which is 68℉. The IR values will be slightly higher than at 72℉ but either temp is fine as long as you stick to a constant value.

I deliberately did not post measured numbers as the measurement of IR is not a simple matter and needs to be surrounded by a few caveats. This particular thread is about a trivial mechanical improvement and a discussion of the relative merits of the meters deserves a more considered one but my comments were based on the following:
  • We know IR is highly temperature dependent so all readings were taken at my “standard” temperature of 20℃. Not only does the IR vary widely with temperature, particularly as it gets colder, but different packs have different curves so there is no universal rule. IR does not vary much with state of charge above 15% or so but it does change as the pack ages even if the pack stays well balanced in both cell voltage and IR.
  • We know the values given by IR meters and IR reporting chargers are very dependent on the method of measurement. In fact IR is not an intrinsic property of a LiPo but a derived value calculated from loading it with a known current pulse and measuring the voltage drop. IR has the value of resistance (Ohms) but is actually a dynamic voltage drop. Accordingly both the current load used and the time allowed for the voltage to drop are very important. The latter is a balancing act between the voltage settling time and the onset of cell voltage drop under continuing load.
  • The critical properties of an IR meter are CONSISTENCY and ACCURACY. Of these consistency is far more important. You want to be able to rely long term and across packs on the values being representative. If they do not agree with another method that is less of an issue since by definition the value of IR is a function of the measurement method.
The Wayne Giles ESR meter has one natural advantage. It measures each cell with the same circuitry. The only variation is the contact resistance at the balance lead which should be unimportant since it is a 4 wire measurement. The other two meters use internal switching circuits to move from one cell to another and these may have variations. *

I have a number of IR meters including a very first Wayne Giles one (hand built using Veroboard!) and a final “universal” version. These two were manufactured probably about 6 years apart and have never been recalibrated. I tested both against the new Revo meter and a Turnigy IR meter.

As test packs I used a brand new Turnigy Graphene 65C 5S 1500mAh pack that has only had 5 break in 1C cycles and a well balanced but old Revo Diamond label 60C 6S 4400mAh from November 2005 that has been kept at storage voltage but had almost no actual use. The former is typical of a modern low IR LiPo and the latter of an aging pack with rising IR which should easily show up any differences.

Results:

1) The two Wayne Giles ESR/IR meters agreed within 0.8 - 1.6% across all cells on the Graphene and 3.0-5.9% on the Revo Diamond. I regard this as excellent consistency considering their age difference. I took the average cell values of the two Giles meters as the benchmark to compare the Revo and Turnigy meters to.

2) The Revo meter reported IR values 4.2 - 7.7% lower on the Graphene and 11.2 -15.2% lower on the Revo Diamond.

3) The Turnigy meter reported IR values 1.2 - 12.9% lower on the Graphene and 20.6 - 34.8% lower on the Revo Diamond. This meter shows much greater variation than the Revo meter.

Hence my comment that, “The Revo meter doesn't give exactly the same values as the ESR/IR meter but it's a lot closer and more consistent than the Turnigy.”

John

* Update: I just looked at the Turnigy vs. the Revo reading times and I think from those that the Turnigy switches between cells while the Revo does a simultaneous measurement on all the cells at the same time as it return IR values almost instantaneously. The microprocessor used can have up to 11x 10 bit AD converters configured and there are 8 OpAmps on the board so this makes sense. Any errors between cells will then be due to different overall AD accuracies.
Last edited by jj604; Jun 27, 2017 at 02:37 AM. Reason: Added Update
Jun 25, 2017, 07:29 AM
Wanted for breaking Ohm's Law
Dennis Sumner's Avatar
I've got an Honesty meter coming Monday. (According to FedEx tracking) I've been using my iCharger 106 to check IR so looking forward to this stand alone meter. I wish Revo would have put it into a case......but m sure case less helps to keep costs down.

Great comparison info John.
Jun 26, 2017, 08:14 PM
Registered User
vollrathd's Avatar
Quote:
Originally Posted by Dennis Sumner
. I wish Revo would have put it into a case......but m sure case less helps to keep costs down.

.
I built up several dozen of these earlier this year. Without access to a CNC milling machine, putting the electronics into a case was a lot more work than expected.

BattIR meter
https://www.rcgroups.com/forums/show...ircuitboard%21
Jun 27, 2017, 04:45 AM
Registered User
Ken Myers's Avatar
vollrathd's DIY Ir meter is an excellent design that uses the 2-Tier DC method to obtain an ohmic value for the battery and its cells internal resistance (Ir). It provides consistent and reliable results. If you have the skills to put one together, you won't be disappointed.

I received my Revo meter yesterday. I'll be working on some comparisons soon for the Wayne Giles meter (16 amp load version), vollrathd DIY meter and the Revo.

Right now, I'm puzzling over what to do about the 'surface charge' when using the single, or typical, DC internal resistance measurement method used by the Wayne Giles and Revo meters. No one has ever taken the surface charge into account, nor the dynamic relaxation effect.
Jun 27, 2017, 09:36 AM
Tim Lampe; Hobbico R&D
KRProton's Avatar
My wife has her own built-in Honesty Meter and uses it on me all the time. She's out of town right now, so I better not screw up, or she'll know!

I apologize for that - couldn't resist.

Tim
Latest blog entry: Latest blog entry
Jun 27, 2017, 10:39 AM
Registered User
Ken Myers's Avatar
I'm just starting to review and test this meter. I just got some interesting readings using my Fluke 115 Multi-meter to read the resting cell voltages and then compare them to the Revo IR meter.

Fluke multi-meter; #1 3.843V, #2 3.842V, #3 3.842V
Revo; #1 3.826V, #2 3.833V, #3 3.832V
Second test right after the first
Fluke multi-meter; 3.843V, #2 3.841V, 3.842V
Revo; #1 3.817V, #2 3.833V, 3.831V

Has anyone else noted this?
I don't want to cut off the XT60 plug if I am going to have to send it back.

Thanks,
Ken

Please see post #13 below for clarification
Last edited by Ken Myers; Jun 29, 2017 at 06:51 AM. Reason: added note
Jun 27, 2017, 10:56 AM
Registered User
Quote:
Originally Posted by Ken Myers
I'm just starting to review and test this meter. I just got some interesting readings using my Fluke 115 Multi-meter to read the resting cell voltages and then compare them to the Revo IR meter.

Fluke multi-meter; #1 3.843V, #2 3.842V, #3 3.842V
Revo; #1 3.826V, #2 3.833V, #3 3.832V
Second test right after the first
Fluke multi-meter; 3.843V, #2 3.841V, 3.842V
Revo; #1 3.817V, #2 3.833V, 3.831V
Ken
I have the original Wayne Giles ESR-IR meter, so I wont be wasting my money on this latest Asian random number-generator.
Thanks for the info - I was thinking of buying it.
Last edited by Zerts; Jun 27, 2017 at 11:08 AM.
Jun 27, 2017, 12:12 PM
Wanted for breaking Ohm's Law
Dennis Sumner's Avatar
Quote:
Originally Posted by Ken Myers
I'm just starting to review and test this meter. I just got some interesting readings using my Fluke 115 Multi-meter to read the resting cell voltages and then compare them to the Revo IR meter.

Fluke multi-meter; #1 3.843V, #2 3.842V, #3 3.842V
Revo; #1 3.826V, #2 3.833V, #3 3.832V
Second test right after the first
Fluke multi-meter; 3.843V, #2 3.841V, 3.842V
Revo; #1 3.817V, #2 3.833V, 3.831V

Has anyone else noted this?
I don't want to cut off the XT60 plug if I am going to have to send it back.

Thanks,
Ken
Ken,

Just checked mine against my Fluke 77.

Revo #1: 3.972 Fluke 77: #1: 3.97
Revo #2: 3.968 Fluke 77: #2: 3.96
Revo #3: 3.985 Fluke 77: #3: 3.98

Looks pretty good......switched over to APP connector after I verified it worked......
Jun 27, 2017, 02:42 PM
Registered User
Ken Myers's Avatar

False Alarm, Sorry


Thanks Denny.

I took some more readings.
In the images below, on the left is the table when I was looking at the voltages to the thousandth place, as measured.
The table on the right of that same image capture shows exactly the same data rounded to the hundredths place.
The actual raw differences are extremely small, and I should have realized that.

What threw me was that I've been using the vollrathd DIY meter for so long now, I was used to seeing much smaller measured differences. I am spoiled with the accuracy of his DIY meter.

The data capture in the image on the right is from the same pack at about the same voltage, but on a different day and it is compared to the vollrath meter.

I really think the comment about this meter being an Asian meter was uncalled for at this moment, because not enough of us have this meter to give a fair opinion based on our evaluation of the product.

I am absolutely not recommending that anyone buy it or NOT buy it. I am very sorry if my original data implied that. That was not my intent. As I said, the percentage differences are tiny, and when looked at as hundredths of a volt, not thousandths, the voltage values are NOT as alarming as they seemed to me at first.
Last edited by Ken Myers; Jun 27, 2017 at 02:50 PM.
Jun 27, 2017, 11:36 PM
Registered User
vollrathd's Avatar

BattIR meter comments


As indicated in the first report in the BattIR thread, the original design used seven voltage dividers with 1% resistors. The tap on the voltage dividers were connected to seven different A/D inputs of the MicroChip microcontroller. The PicChip has 1^12 or 1/4096 resolution, and that PicChip tracked digit per digit from zero to full scale, as compared to my Fluke 87V, accurate to 0.05% on DC.

Problem was, to get individual cell voltages, as an example, the software measured the total voltage of all 7 cells, and subtracted the voltage of all 6 cells to get the voltage for the seventh cell. And, that cut the resolution of the voltage measured down. The full scale voltage resolution about was 40 VDC divided by 4096 or about 0.010 Volts DC. Add to that, the 1% resistors on a 40 Volt full scale range is 0.4 Volts DC!! That 0.010 volt resolution added another problem. On high power LiPo battery packs, their voltage doesn't drop much under the 15 Amp load applied by the BattIR meter. And, you need significant voltage resolution to accurately calculate the battery IR values.

Going to the big pile of electronic relays allowed a single PicChip channel to measure each of the cell voltages on a 5 Volt DC full scale, increasing the resolution by a factor of 10. Plus, no precision resistors are required. And, there were no issues with slight variations between the A/D converters on the PicChip. Adding a number of voltage readings per cell also improved any "jitter" of the voltage reading to less than one digit. And, the multiple readings did simulate a little better than 1/4096 voltage resolution.

The final circuit design uses a precision 5 VDC linear regulator for the voltage reference. The PicChip does have an internal voltage reference, but using it would have limited the circuit to a maximum of 4.000 VDC full scale. If I were to build any more of these units, I'd add a simple 10 turn trimmer resistor in the voltage measuring circuit to make it easy to precisely adjust the BattIR meter to match a precision DC voltmeter, such as the Fluke 87V.

Of the dozen of my club members that bought one of the BattIR meters, one did want his tweeked to exactly match his Fluke 87V. I did it by adjusting the software calibration point. His meter now tracks his Fluke to 0.1%.
Jun 27, 2017, 11:45 PM
Registered User
vollrathd's Avatar
Quote:
Originally Posted by Ken Myers
Right now, I'm puzzling over what to do about the 'surface charge' when using the single, or typical, DC internal resistance measurement method used by the Wayne Giles and Revo meters. No one has ever taken the surface charge into account, nor the dynamic relaxation effect.
I'm kind of curious about that "Surface Charge" on a LiPo battery pack, having run across it during the design stages of the BattIR meter. What does it take to eliminate it or reduce it? Would a three second or so load of 5 Ohms on the battery pack make much difference on the surface charge?

A123 battery packs are a different animal, and that surface charge thing doesn't have a great effect on it.

Just wondering.

PS
I've got a 12S3P 7500 Mah A123 pack on my giant scale model with the Rimfire 50 cc electric motor. Brand new, with the same prop, the motor tached at 7200 RPM. After 165 flights on that pack, the Rimfire motor now tachs at 7170 RPM, a drop of 30 RPM.


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion CC3D Revo no motor using TX stick marco048 Beginner Multirotor Drones 4 Mar 06, 2017 05:00 AM
Discussion Timing accuracy / honesty tom43004 Hand Launch 76 Jul 31, 2016 01:23 AM
Discussion Should I use PPM with Devo 10 (deviation) and CC3D (eventually Revo) n8huntsman Beginner Multirotor Drones 1 Mar 25, 2016 10:16 PM
Discussion Traxxas Mini Revo Battery Compartment Mod For Larger Lipos KevinMillard Off-road Cars 0 Jan 31, 2016 07:10 AM
Discussion I can respect his honesty logan5 Life, The Universe, and Politics 84 Nov 03, 2015 09:55 PM