Fluke 233 batteries flat in 4 months - Unbelievable response from Fluke!

[Zad]

"Yes sir, it's designed to do that" - Well, not in so many words, but that's what it boils down to.

Regulars may remember my mini-rant about my Fluke 233 meter. That's the one whose head unit can be removed to remotely monitor the measured quantity. One of the features they don't mention though, is that even if you don't use it at all, the batteries will run flat within 4 months. Both base and display units (which have a total of 5 AA cells) draw around 1mA even when both units are OFF.

This week, finally spurred on by reports both on here and on Element-14's site ( http://www.element-14.com/community/thread/7523?tstart=0 ) I decided I should approach Fluke and ask if this was intended to be normal, and if anything could be done about it. Here is the response that got posted to the Element-14 forum (I have had no personal email reply yet). It is a public forum, so I feel able to re-post here. I have added paragraphs and spacing to make it readable, as the original poster seems not to know how to do so;

Hello Viewers, The 233 is unique in its design and implementation with its wireless display utilizing both iR and RF communication between the meter base and the remote display. While docked the meter utilizes iR communication and when the display and base are separated RF communication is used.

When new the meter is shipped with paper tabs inserted into the battery compartment for both the base and the display to keep the batteries disconnected from the meters power supplies. This is to insure that the batteries have a full charge and the customer gets the longest battery life possible. The reason for a short battery shelf life is that when the meter is turned off and the display is mated, the iR communication is still active. This was done by design to give the user the most flexibility when using their 233 meter.

The design allows the user to power on and to wake up the meter while the display and body are disconnected. It simplifies the meters operation. This use model was important to customers during our initial research for the 233. The alternative would have been to have two separate power buttons, one for the base and one for the display. Our research indicated that this was a less attractive solution to our implementation of one power up option for both parts.

The trade off is the unit consumes more power but gives the users more flexibility. If the meter were to go into battery save mode and the iR communication were not active, the display and meter body would need to be re-connected before the display could power up. This option and two separate power up processes were not attractive in our research.

To optimize the battery life, for intermittent use or extended periods of storage, it is recommended that the paper tabs be re-inserted into the battery compartments or the batteries be removed completely.

Jim Newell Sales & Application Engineer Fluke Corporation

(My bold emphasis)

I'm stunned! Not only do they see this battery life as acceptable and what their customers asked for, they actually want me to take the batteries out or insert fiddly paper tabs when I think I may not be using the unit (unfortunately I don't have foresight so am unable to comply with this).

I am going to make some recommendations:

1) As a short term fix, that the base unit only transmits maybe a couple of times a second, thus hugely reducing the average current load needed to constantly transmit. Additionally, reduce the IR LED output power by increasing the value of the series current limiting resistor.

2) That their next design revision uses either a mechanical or magnetic switch to detect when the display unit is undocked. It could even have electrical contacts in the docking rails that are shorted out and are detected as open circuit when undocked.

3) That the micro in the display be programmed to react to in interrupt generated by the IR LED rather than being permanently active (as it appears to be).

4) If they can't get it sorted then they should supply rechargeable batteries and a charger, or integrate one.

Maybe Fluke should employ fewer sales and marketing people and 1 or 2 more design engineers? Jeez...

Mike

[saturation]

Ouch. Sorry to hear; first models often have bugs so I avoid anything that is 'new' unless its an old model with some incremental changes. 

Recall the Fluke 85 series III has a bug with power consumption in sleep mode, it never goes to sleep even if it appears to be, but no one caught it so it was never fixed in that series; an eevblog user noted it and complained, I think they gave him a new Fluke 87V, years after he bought the original. 

[Psi]

Didn't Dave review/rant about this specific issue in a video already?

[ziq8tsi]

2) That their next design revision uses either a mechanical or magnetic switch to detect when the display unit is undocked. It could even have electrical contacts in the docking rails that are shorted out and are detected as open circuit when undocked.

The base unit certainly should not have any exposed contacts.  They could be at dangerous potential when the meter was in circuit.

4) If they can't get it sorted then they should supply rechargeable batteries and a charger, or integrate one.

Really?  That would make the meter even more expensive, and the initial purchase cost already dwarfs its battery costs.  Plus, rechargeable batteries are hardly the solution that springs to mind when designing a product to be left unused for months.

I think their explanation makes a lot of sense.  It is more convenient and intuitive for the power state to be communicated continuously between the units.  And most people buying a high-end meter will be using it regularly.

I do agree, however, that even with the selected design, it does not sound too difficult to decrease the average standby current one more magnitude to 0.1mA -- and then the problem would go away.

[Zad]

2) That their next design revision uses either a mechanical or magnetic switch to detect when the display unit is undocked. It could even have electrical contacts in the docking rails that are shorted out and are detected as open circuit when undocked.

The base unit certainly should not have any exposed contacts.  They could be at dangerous potential when the meter was in circuit.

They needn't be exposed, but integrated into the runners. They wouldn't be anywhere near the meter PCB or conductors, and would only exist to provide a short for the display head to measure (difficult to explain without a diagram). I do prefer the switch or magnetic method though.

[mikeselectricstuff]

It does not take 1mA to keep an IR link between a closely coupled IR pair active. It isn't rocket science.
This is just lazy design and not acceptable from a quality brand like Fluke.

[Kiriakos-GR]

This is the second thread about the same topic , and we had all ready spent lots of ink about it .. ( in the first one )

There is plus and minus in this special design , and this design does not work for every one as he likes.

Fluke had researched the market , so to offer one solution that works for the most people with special needs.
If some one does not belong in a such target group , then he has an single option, to sell it and get something different.

The same suggestion goes to any one who gets something, that later on looks not capable to act as expected ,
sell it , and get the right one for you, if it exists .      

[Zad]

Whooooosshhhh....

(That was the sound of an idea flying over someone's head at high speed)

The fact that a particular functionality is added does not excuse poor design and severe compromise in another area. In this instance, the functionality is the ability to switch the remote display unit on or off from the base unit at any time, whether docked or undocked, and to activate the RF link when undocked. The compromise in this instance is 1mA current with the IR LED transmitting ALL the time.

My contention is twofold:

1) Even retaining this hardware, it would be possible to reduce the average consumption to below 0.1mA without in any way compromising any functionality. Thus extending battery life to a more reasonable 2.5 years.

2) By designing the unit properly in the first place, it would have been possible to reduce the current consumption to zero. Again without in any way compromising any functionality Nor would it have cost any more to design or manufacture, and there would be no reduction in reliability.

I'm sorry if a second posting on the subjects offends your sensibilities Kiriakos, but the subject relates to the reply from Fluke, rather than the original problem itself. If you don't like it, don't waste time reading it.

[mikeselectricstuff]

Whooooosshhhh....

(That was the sound of an idea flying over someone's head at high speed)

The fact that a particular functionality is added does not excuse poor design and severe compromise in another area. In this instance, the functionality is the ability to switch the remote display unit on or off from the base unit at any time, whether docked or undocked, and to activate the RF link when undocked. The compromise in this instance is 1mA current with the IR LED transmitting ALL the time.

My contention is twofold:

1) Even retaining this hardware, it would be possible to reduce the average consumption to below 0.1mA without in any way compromising any functionality. Thus extending battery life to a more reasonable 2.5 years.

2) By designing the unit properly in the first place, it would have been possible to reduce the current consumption to zero. Again without in any way compromising any functionality Nor would it have cost any more to design or manufacture, and there would be no reduction in reliability.

I'm sorry if a second posting on the subjects offends your sensibilities Kiriakos, but the subject relates to the reply from Fluke, rather than the original problem itself. If you don't like it, don't waste time reading it.

Totally agree - remote on/off via RF would draw significant current (although still probably not 1mA) and the only choice the competent designer really has is the balance between battery drain at each end.
However this meter will spend most of its life docked, so off mode power draw when docked is by far the most important parameter. On/off sense via IR on a properly designed closely coupled IR tx/rx  is pretty simple. A simple phototransistor will easily pull down a CMOS input cleanly to wake an MCU with no more than a pullup resistor, so there's no static IR receive amplifier current draw. If on/off from either unit is required, it only takes a very brief comms handshake, maybe 100uS once a second, so  10mA IR LED current and some CPU wakeage would struggle to need more than 100uA avarage without really trying, probably a lot less.
It is poor design, pure and simple, and a shame that an otherwise unique & interesting product has been spoiled by it. 

[Kiriakos-GR]

Whooooosshhhh....

(That was the sound of an idea flying over someone's head at high speed)

bla bla bla ..

The fact that a particular functionality is added does not excuse poor design and severe compromise in another area.

bla bla bla ..

I'm sorry if a second posting on the subjects offends your sensibilities Kiriakos, but the subject relates to the reply from Fluke, rather than the original problem itself. If you don't like it, don't waste time reading it.

My apologies from my public reply , that does not belong in the ones that you would prefer,
but I have the feeling , that you use this minor issue of the DMM, as excuse to blast the door and raise your voice.

Technically  , possibly it is room for improvement , but any suggestion should be followed with some responsibility.

It looks that you have use and other forums , so to publicly protest about the reason that you think as  " poor design and severe compromise ",
but with out any true gain .
And so here you are again , so to set an new fire and keep the flames alive.


Well if it does enjoy you , keep doing it.

But have this is mind too , any reasonable person, he would just remove the battery ,
if he had set it in storage , and that's it.

My current impression are that you got the wrong item , for your needs, what ever those are.   


 

[Wim_L]

However this meter will spend most of its life docked, so off mode power draw when docked is by far the most important parameter.

If I had to make a guess, their target market for this meter would have been a repair technician who is using the meter all day long and making much use of the detachable screen (otherwise, surely said technician would get a cheaper meter with integrated screen). That would mean the drawback you identified wouldn't be so significant, and that your needs might be better served by something else. Or maybe the way you use it is uncommon enough that nobody ever made a meter optimised for your style of work.

[CodeDog]

I bought my 233 a couple of years ago and finally got tired off changing the batteries every few months.

Under its 3-yr warranty, Rapid-Tech Equipment (Fluke agent I bought it from) got it fixed ... new hardware (boards in the main and remote) and new software (v2.0). And they also sent back its test results (all pass).

Just got it back today and couldn't be happier ... so here's a shout out of thanks to Rapid Tech (and Fluke) .

cheers,

[EEVblog]

So what was the fix?

[lewis]

I stopped using my 233 too, it was just too much of a pain in the arse.

I'd be very interested in knowing what the fix was too!

[CodeDog]

I don't know specifically what the cure is. They swapped out boards from both the main body and the detachable head, and also upgraded the firmware. Apparently the fix would cost about $200  + GST if it is done out of warranty. I was without my meter for about 6 weeks, since apparently parts had to be shipped in.

If I were to guess, I would say that the firmware upgrade kills the comms between body and head when the unit is off, and the hardware upgrade allows the head to wake up when the switch on the body is moved. The head has to be docked to wake up. Not sure if this was the case previously, but I suspect that the head could be woken if undocked before.

I'll measure the current drain in off mode in the coming days.

cheers,

[Zad]

If you feel able, it would be interesting to see a photo of the PCB to compare with mine. At the moment my fix is to leave it on the shelf and use my early 90s vintage Fluke 77.

[SArepairman]

how about a copper foil-insulator-copper foil shunt that connects to a microswitch to the batteires with parallel 30 gauge wires (for current capability and thinness)? it would require some case work but you could probably do it with a bit of dremel milling, those fluke housings are pretty thick. or just make traces that run along the battery compartment top out of copper tape that leads to a switch which is inserted into a small cavity dug out with a dremel

are yoou man enough to perform brutal surgery on your expensive fluke>

[JoeO]

Fluke needs to be concerned about their designs, not the color of imported meters.

[sangvikh]

Did you get around to measure the current CodeDog?
I use this fluke for work. The remote display is really handy for measuring current while changing parameters in a plc.
I use it a lot, so not the powerconsumption is not that big a drawback, but it is dead after 3 months of no use.

[H.O]

Hadn't noticed this thread before but I've had a 233 at work for the last 10 months. It's seen very little use, mostly sitting in my drawer obviously turned off. It's still on its first set of batteries.

Measuring the battery voltage with the unit turned on and in DCV-mode shows 4.36V across the three cells in the main unit and 1.49V across the two cells in the display unit.

The meter works fine and does not show a low battery indicator but I would say that the two cells in the display unit is soon in need of a Batterizer :-) So if there's been a change to the hardware I probably have the later model. I can open it up and take some photos if needed.

[sangvikh]

I got mine in 2013, so they have probably fixed this issue. Too bad I'm stuck with this lemon.
I will not pay 200 dollars for an upgrade as someone here said it would cost, might as well buy a new one then.