Alternative to DR1000 butchery for a Power Switch

[curbarthedog]

Controversial I know, but I actually quite like the capacitive switches but agree the biggest problem appears to be power drain. Hence a few people are investigating home brew modifications.

Now I am no software or hardware engineer but have a suggestion to exploit an extremely annoying "feature" of my DR1000 to achieve the same goal.

Whenever I plug my DR1000 in to charge, it turns itself on. In other words, applying power to the mini-usb port turns the whole thing on.

What if...the capacitive buttons were completely disabled when the device is switched off and you switch the device on via a "dongle" of some kind i.e. mini-usb plug with a small battery?

Granted, this would increase insertions into what would appear to be a fragile USB port but could this be a practical solution?

If this were possible, an option could be added to the settings to "turn on via dongle" or "turn on via buttons" to allow people the flexibility.

Just a though!

p.s. if iRex are reading this, please issue a patch for the lock ups when using the tabs, the device is currently impossible to use for a business user!

[Shaggy]

Originally Posted by curbarthedog

What if...the capacitive buttons were completely disabled when the device is switched off and you switch the device on via a "dongle" of some kind i.e. mini-usb plug with a small battery?

It's a good idea. How would you do it though? It sounds like opening the unit generally causes significant damage. Is there a way to disable the switches, keep the "power on" via usb intact, without permanently damaging the device?

It would be a cool modification if you could.

Of course, this is assuming that the buttons are a major contributor to the power drain, which I don't think anybody has really proven yet.

[Kent Walters]

Here, I said:

Originally Posted by Kent Walters

... To use the cap sense switch to turn the device ‘on’ requires the circuitry to be always powered up--even while the device is ‘off’--with it’s clock running...

And

On my ‘experimental’ (broken screen) DR I have noticed that, when connecting the battery, the DR behaves exactly the same as when power is applied thru USB, and after being unplugged for a week, the battery was still fully charged; a power switch on the battery would have eliminated power drain while off, whether the drain was caused by the cap sense circuitry, a firmware bug or anything else.

And

Meanwhile, I’m installing a battery door and switch on my good 1000S.

If a full power down can be arranged thru software/firmware (and I don’t know), then curbarthedog’s USB suggestion is excellent, in that it provides no risk to the DR or the warranty. It could be done using a USB power pack, adding the advantage of extended runtime.

This weekend (hopefully) I will void my warranty and install a battery door on my good DR, just because I’m tired of continually charging it. I will also install a switch, but I’m having trouble coming up with a switch design that could be (relativly) easily implemented by others, who I presume do not have the machinery and tooling that I have. My goal is a switch that appears factory installed, and can be installed thru the battery door opening--a challenge, given the case’s design.

At some point warranties will expire, and/or the DR’s battery will reach the end of it’s life, and need replacement, anyway. I don’t know how many owners would want to send their DR’s to iRex; I will guess that iRex will charge something in the $75 to $100 range for a replacement, and likely take about two weeks (just guessing, mind you).

And Shaggy, power drain while ‘off’ is a fact, regardless of it’s cause. The cap sense switches must be powered in order to work, even just to turn the DR 'on.' My unplugged battery held it’s charge for a week, and was still fully charged.

Kent Walters

[Shaggy]

Originally Posted by Kent Walters

And Shaggy, power drain while ‘off’ is a fact, regardless of it’s cause.

Of course it is. But many people are making assumptions about the cause, and trying to treat them as fact.

My point all along has been that we shouldn't jump to any conclusions when we have no data to support it.

BTW, did you know that there is at least one known bug in the software that will also cause a high power drain while "off"?

[Kent Walters]

I say again, the cause is irrelevant. Whether it's a bug or hardware, the DR will not hold a charge, and the cap sense switches must, by their design, be a contributing factor, and whether they are the main factor or not means squat.

No one will be happier than me when iRex, or anyone else, fixes the bugs in the DR. Instead of just waiting for that, though, I'm going to do whatever I can to make my DR more suitable to my needs and desires.

I can't let iRex off the hook for bad design by excusing poor performance with the excuse of a bug that iRex made--There's no logic there.

Kent Walters

[Shaggy]

Originally Posted by Kent Walters

I say again, the cause is irrelevant. Whether it's a bug or hardware, the DR will not hold a charge, and the cap sense switches must, by their design, be a contributing factor, and whether they are the main factor or not means squat.

It does make a difference. If the buttons are not a major factor, then curbarthedog's suggestion will not be effective.

Your suggestion about installing a switch to completely cut all power to the device will work, but that's not what curbarthedog was talking about above.

Knowing the cause is important so that you know what you need to do in order to fix the problem.

I can't let iRex off the hook for bad design by excusing poor performance with the excuse of a bug that iRex made--There's no logic there.

The opinion that it's a bad design is based on the assumption that a software bug is not the main reason for the poor performance. Is the performance problem a design problem, a software problem, or both? Yes, you can fix the problem with a design change, but that change may not even be necessary if the cause is a software problem (which we already know exists).

If a software fix takes care of the poor performance, then fewer people would be of the opinion that the buttons are a bad design choice.

My only point to this is that when you're discussing hacks to the device which are likely to void the warranty, it's a good idea to be clear about what really needs to be done in order to get the results you are looking for. You may be OK with voiding a warranty unnecessarily, but many others will not think that's such a good idea. IMO, taking the time to actually find out if the buttons are a significant part of the problem would be beneficial if someone is considering this modification.

[Kent Walters]

Originally Posted by Shaggy

The opinion that it's a bad design is based on the assumption that a software bug is not the main reason for the poor performance. ...

Oh, man; You keep crediting me with ‘assumptions’ of your own creation; I have made no assumptions--let me summarize why I believe that the cap sense switches, as designed and implemented on Irex’s DR1000S, were a bad design choice:

1) (This is subjective, I know) To design a battery powered device so that it will always draw power, even when turned ‘off,’ is (IMHO) Dumb. To do this on a device that is very power hungry to begin with, and arguably underpowered, is absolutely absurd. To do such a thing contradicts anything I have ever seen in any ‘Best Prectices’ manual.

By design, the cap sense switches draw power while waiting to be touched, and, as implemented on the DR, that is whether the DR is ‘on’ or ‘off.’ Whether this is the major cause of drain or not, is, to me, a moot point; it is draining the battery, and I want to stop it.

This could have been totaly circumvented by the simple addition of an on/off switch by iRex. This would prevent power drain while ‘off,’ no matter what the cause is.

And 2) As implemented, the DR’s cap sense switches are tempermental, funtioning inconsistantly and unreliably. Sometimes they respond to a finger or hand just coming too near, and sometimes they don’t respond to an actual touch. Others have reported interferance from florescent lights, and I find that, sometimes, when working at my computer, I have to turn off the computer’s Bluetooth. I always get better performance at my couch or recliner than at my computer, but I often need to work at my computer. I will ask anyone to compare this to the functioning of a Touch Wheel iPod, which is also cap sense technology, and seems to function reliably, and as expected, no matter the environmental field variables (RF/EMI, Biorythms or phases of the Moon). It might also be noted that iPods do fully power down when ‘off.’ (Good design)

If iRex could not design better circuitry, they should have stuck with conventional switches--the device would be much more reliable, and even pleasurable to use, had they used old-school switches. Yes, again, this is subjective; who cares?

The effectivness of curbarthedog’s suggestion depends on the ability to power down the DR thru software, and the bug you refer to suggests that that is not possible at this time. His idea is excellent; anyone could do it, and without voiding warranties. However, if the bug you mention is responsible for the majority of battery drain while off, and that bug is fixed, the point is moot; I simply have no confidince that irex will fix it. If the switches are causing a significant drain, and the switches can be truly turned off, curbarthedog’s suggestion is, I believe, the absolute best way to go--for everybody. While I wait for others to do their things, though, I will install a switch.

I have nowhere made a claim that a software bug was not the major factor in power drain, as I addressed where you first quoted me above. Frankly, I don’t care what iRex has done to ‘cause’ the battery problems; iRex has sold a product with serious faults, and some of those faults were by their choice and design. I can forgive ‘bugs,’ but while I wait for iRex to fix them, I’m wasting time, money, electricity, and my battery’s life. I have spent well over $1000 on this thing so far, and it is not as reliable as a free cell-phone. You might add to my costs the fact that I bought my EZReader only so that I could enjoy the benifits of ePaper without the annoyances of the DR, on those occasions when I just want to read.

I have noted the potential of voiding warranties, and not advised anyone to do so. Each person makes their own choice, and for those who choose to hack, I offer my help, knowledge and experience.

I have determined, to my own satisfaction, that the DR (as it is today) would greatly benifit from a power off switch. While you might choose to quibble about exactly what causes the drain, or which cause is most culpable, I will solve my problem, and install the switch.

Having solved the drain while ‘off’ problem, I will use my meters, logic probes, O’scope and computer to address the possibility of replacing the cap sense switches; you can monitor the swiches power usage, if you choose; I don’t see any benifit.

Kent Walters

[curbarthedog]

Shaggy,

I agree we are speculating, but in the absence of any kind of information, assistance, forecast issue fix dates that is all we can do.

For my suggestion to work, it would need someone to execute via software or firmaware the disabling and turning off of the cap sensors.

Unfortunately, I suspect this may need iRex to action and as action does not seem to be a word in their dictionary, I suspect this is an academic exercise.

I suspect the unreliable performance of the cap buttons relates to the PDF reader implementation. On some docs it is quick, on some it is woeful. There is even a difference within the documents! Hopefully, the new Adobe PDF implementation will make a difference.

[Shaggy]

Originally Posted by Kent Walters

1) (This is subjective, I know)

Yes.

By design, the cap sense switches draw power while waiting to be touched, and, as implemented on the DR, that is whether the DR is ‘on’ or ‘off.’ Whether this is the major cause of drain or not, is, to me, a moot point; it is draining the battery, and I want to stop it.

IF the drain is insignificant, then why would you care? Most small electronic devices don't really turn "off" these days. As long as the capacitive buttons are not a major source of power loss (which we don't know), then does it really matter? At that point, it's just a subjective preference of whether you like capacitive buttons or not. Obviously, you do not.

If they are a major source, then I absolutely agree that it was a terrible design decision. I just don't think we really know enough to make that determination yet.

[Shaggy]

Originally Posted by curbarthedog

I agree we are speculating, but in the absence of any kind of information, assistance, forecast issue fix dates that is all we can do.

Something to keep in mind that may be encouraging... as you know, they are planning on replacing the PDF reader application on the device. Right now it's an open source implementation and they are changing that to a version of Adobe. The existing software bug that is known to cause major power loss seems to be related to the reader application.

I don't know if switching the PDF reader will solve that software bug, but it is definitely reasonable to think it might.

For my suggestion to work, it would need someone to execute via software or firmaware the disabling and turning off of the cap sensors.

Can the sensors be turned off via software/firmware?

[curbarthedog]

Well, that I guess is the $64,000 question!

Anyone have any insights?

[Shaggy]

Originally Posted by curbarthedog

Well, that I guess is the $64,000 question!

Anyone have any insights?

The code is open source. If the ability exists, we don't necessarily need iRex to do it. You'd just need one of the community developers to take a look and see if that functionality is there.

My gut feeling is that the sensors are not able to be shut off via software, but I don't really know.

[Kent Walters]

curbarthedog;

The cap sense switches give me problems with other formats as well, not just PDFs. Some days are better than others, no matter what I’m doing... the behavior seems related to whatever EMI fields are at play, but behaviour is inconsistant even out in the open; it might even be related to changes my own physiology/capacitive field--I don’t know. I do know that I don’t have these problems with any other devices (like iPods), and it seems to me that the DR’s circuitry is just too sensitive.

I have a suspicion that Adobe’s focus will be more toward DRM than users, but I hope that eventually eBook readers will be able to take advantage of all that the PDF format has to offer--not holding my breath, though.

Shaggy,

So, do you disagree with my subjective opinion that designing a battery powered device that will never power off is a bad choice? Or do you believe that it is a good idea to have a constant drain on a battery, even while the device is ‘off?’ Or is it that you just object to someone stating an opinion? Hmm...

Originally Posted by Shaggy

IF the drain is insignificant, then why would you care?

My DR is the only battery powered device that I have, or even know of, that will not operate after three or four days of being turned ‘off,’ and that is significant!

Originally Posted by Shaggy

Most small electronic devices don't really turn "off" these days.

You have changed the subject here; I was referring only to battery powered devices; I will ask you to name another one that performs this way by design; a flashlight, maybe? Even a cell-phone can be turned fully ‘off.’

Originally Posted by Shaggy

As long as the capacitive buttons are not a major source of power loss (which we don't know), then does it really matter?

I have said repeatedly that it doesn’t matter what is the ‘major source of power loss’ is, only that there is power loss; the ‘major’ cause is irrelevant.

Originally Posted by Shaggy

At that point, it's just a subjective preference of whether you like capacitive buttons or not. Obviously, you do not.

Obviously? I cited the iPod cap sense wheels because I have had two of them just so equipped; I loved ‘em both. They always worked as intended, and exactly as I expected--unlike the DR--and regardless of the environmental variables; whether I was on the beach or sitting at my computer. I like cap sense technology, and have been experimenting and playing with the technology for years--give me a break. It is iRex’s implementation that sucks; if they couldn’t do it right, they shouldn’t have done it at all, period.

Originally Posted by Shaggy

If they are a major source [of power drain-kw], then I absolutely agree that it was a terrible design decision. I just don't think we really know enough to make that determination yet.

The ‘terrible design decision’ was not the choice to use cap sense switches per se, it was choosing to use a poor implimentation of them, and to not provide a way to turn the DR ‘off’ to allow it to remain charged between uses--why is it so hard for you to understand what I’m actually saying?

Kent Walters

[Mackx]

Originally Posted by Shaggy

The code is open source. If the ability exists, we don't necessarily need iRex to do it. You'd just need one of the community developers to take a look and see if that functionality is there.

My gut feeling is that the sensors are not able to be shut off via software, but I don't really know.

I do not think that the keys and leds are directly controlled by the main processor (ARM) but by an extra processor (see kbdupgrade-component). The code of that processor is not published, so it will not be easy to change it. This processor MIGHT have the ability to powerdown the sensors (and itself?).

[tjdean256]

I guess it boils down again to whether you feel the device is useful or not, regardless of what an individual may consider design flaws. To me, the DR1000s is extremely useful. I get a lot of work done with it. What flaws or bugs it may have are easily worked around, in my opinion. If you can design a better one, do it.

td