HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Nov 30, 2005, 08:10 PM
know it all
Miami, Fl
Joined Nov 2003
2,949 Posts
there is no way to do that with just a chip upgrade.. there is no way to confirm anything, just a pot that adjust current and I dont think the cpu chip can sense that pot.. it just drives the power converter side of the charger. The design would have to be modified with a pushbutton of some sort.

I would love to have that pushbutton, also if it would act like a reset so you dont have to disconnect the power leads of the charger..

Herm

Quote:
Originally Posted by Dan Baldwin
Again, what would be the problem with having the 109 work exactly as it does now, except that it would stop if it increases the cell count, and wait for confirmation by the user?

Dan
hermperez is offline Find More Posts by hermperez
Reply With Quote
Sign up now
to remove ads between posts
Old Nov 30, 2005, 09:15 PM
Registered User
Dan Baldwin's Avatar
United States, CA, Norwalk
Joined Apr 2004
2,728 Posts
Quote:
Originally Posted by hermperez
there is no way to do that with just a chip upgrade.. there is no way to confirm anything, just a pot that adjust current and I dont think the cpu chip can sense that pot.. it just drives the power converter side of the charger. The design would have to be modified with a pushbutton of some sort.

I would love to have that pushbutton, also if it would act like a reset so you dont have to disconnect the power leads of the charger..

Herm
The microcontroller can't directly read the pot, but it can see the charging current. All that it would be necessary for the microcontroller to do to implement what I describe is to turn on the charge current for a brief period of time, check charge current, then turn it back off, beep, and test again. Normal charge would not proceed until the pot has been turned down to minimum for a period of time;perhaps 5 seconds. Duty cycle could be extremely low, so there is virtually no chance of overcharging the pack in the mean time. This cycle would continue for perhaps 5 minutes, and if the user doesn't take any action, the charger would error out and stop.

I think it would be a better charger with a button too, but I'd love to see this change made without using a button so that people can upgrade old chargers.

Dan
Dan Baldwin is offline Find More Posts by Dan Baldwin
Reply With Quote
Old Nov 30, 2005, 09:21 PM
Suspended Account
USA, FL, Apopka
Joined Dec 2000
4,072 Posts
Quote:
Originally Posted by hermperez
there is no way to do that with just a chip upgrade.. there is no way to confirm anything, just a pot that adjust current and I dont think the cpu chip can sense that pot.. it just drives the power converter side of the charger. Herm
Dan that means the CPU would not be able to know the pot was turned off.

Which still means that my proposal in post #137 would work.

We know the microprocessor does control the ability to turn the current on and off.

stage 1; 50% on, 50% off
stage 2: 59 seconds on, 1 second off
stage 3: decreasing time on

Concept 1: If possible develop an algorithm for stage 1 that adjusts the on/off time based on the estimated number of cells that would not bump the cell count on a fully charged pack.

Concept 2: Turn off any current flow in stage 1 for packs under 4S.

Concept 3: Go to stage 2 immediately for any pack under 4S.

I'm sure there are other posibilities, but if the microprocessor cannot sense the pot the idea of a user confirmation is impossible.

Brad
bradpaul is offline Find More Posts by bradpaul
Reply With Quote  (Disabled)
Old Nov 30, 2005, 10:38 PM
Registered User
Walled Lake, MI, USA
Joined Feb 2000
11,182 Posts
Brad, now that's creative thinking. I like it.
Dave Hederich is offline Find More Posts by Dave Hederich
Reply With Quote
Old Dec 01, 2005, 12:35 AM
know it all
Miami, Fl
Joined Nov 2003
2,949 Posts
If it can see the charge current (just because it is displayed does not mean that it knows the quantity..perhaps that is a different chip) then it can be implemented with a new program.. the PIC cpu would have to have two A/D converters.. one for the important voltage and one for the current. Anyone knows what model PIC is used?.. also assuming there is enough room for the new program.

Herm

Quote:
Originally Posted by Dan Baldwin
The microcontroller can't directly read the pot, but it can see the charging current.
Dan
hermperez is offline Find More Posts by hermperez
Reply With Quote
Old Dec 01, 2005, 09:58 AM
Registered User
Walled Lake, MI, USA
Joined Feb 2000
11,182 Posts
Requiring the user to monitor an automatic process and halt the process if it is incorrect is not intuitive for most people, as this is not a common situation with modern consumer electronics devices. The common assumption is that modern consumer electronics devices are designed for intuitive operation and with automatic protection to minimize human error.

The key point to me is that any Li charger with any automatic features must be designed so that it automatically defaults to the safest setting if there is no user intervention. This is not the case with the AF 109 if the autodetect feature can change cell count and continue into an over-voltage condition in the absence of user intervention.

Therefore, I do not see this issue going away.
Dave Hederich is offline Find More Posts by Dave Hederich
Reply With Quote
Old Dec 01, 2005, 10:19 AM
Registered User
rcmann's Avatar
Joined Sep 2003
817 Posts
With this serious problem, why do people buy the 109, when there are so many other good chargers available?
rcmann is offline Find More Posts by rcmann
Reply With Quote
Old Dec 01, 2005, 10:40 AM
Registered User
Dan Baldwin's Avatar
United States, CA, Norwalk
Joined Apr 2004
2,728 Posts
Quote:
Originally Posted by rcmann
With this serious problem, why do people buy the 109, when there are so many other good chargers available?
For people with large packs, the 109 is by far the most bang for the buck. It is capable of charging 9 cell packs, and it is capable of charging up to 8 amps, although not at the same time. If I had to charge big packs, I would probably buy a 109, but since I know it's tendencies, I would take the proper precautions every time I used it.

I think most people buy the 109 because they know and trust Astro Flight, and they want to buy the best and safest. In a thread a while back, one guy posted (paraphrasing)" I'm glad I don't have to worry about lipo fires. I use an Astro 109".

Dan
Dan Baldwin is offline Find More Posts by Dan Baldwin
Reply With Quote
Old Dec 01, 2005, 10:42 AM
Registered User
Dan Baldwin's Avatar
United States, CA, Norwalk
Joined Apr 2004
2,728 Posts
Quote:
Originally Posted by Dave Hederich
Requiring the user to monitor an automatic process and halt the process if it is incorrect is not intuitive for most people, as this is not a common situation with modern consumer electronics devices. The common assumption is that modern consumer electronics devices are designed for intuitive operation and with automatic protection to minimize human error.

The key point to me is that any Li charger with any automatic features must be designed so that it automatically defaults to the safest setting if there is no user intervention. This is not the case with the AF 109 if the autodetect feature can change cell count and continue into an over-voltage condition in the absence of user intervention.

Therefore, I do not see this issue going away.
Hallelujah

Dave, can I consider you a convert?

Dan
Dan Baldwin is offline Find More Posts by Dan Baldwin
Reply With Quote
Old Dec 01, 2005, 11:00 AM
Registered User
Dan Baldwin's Avatar
United States, CA, Norwalk
Joined Apr 2004
2,728 Posts
Quote:
Originally Posted by hermperez
If it can see the charge current (just because it is displayed does not mean that it knows the quantity..perhaps that is a different chip) then it can be implemented with a new program.. the PIC cpu would have to have two A/D converters.. one for the important voltage and one for the current. Anyone knows what model PIC is used?.. also assuming there is enough room for the new program.

Herm
I don't know which PIC Bob is using in the 109, but as I recall it either has a 10 bit or 12 bit A/D built in. The larger PICs have an A/D converter capable of reading as many as 11 analog signals. The analog value read from a port is placed into a 16 bit register (2 8 bit registers), and it would only take a few instructions to check that value to make sure it was below a certain level to make sure the pot has been turned down.

Dan
Dan Baldwin is offline Find More Posts by Dan Baldwin
Reply With Quote
Old Dec 01, 2005, 11:06 AM
**I'm Battman**
RCBABBEL's Avatar
Twin Falls, Idaho
Joined Jan 2005
8,691 Posts
Quote:
Originally Posted by Dave Hederich
Requiring the user to monitor an automatic process and halt the process if it is incorrect is not intuitive for most people, as this is not a common situation with modern consumer electronics devices. The common assumption is that modern consumer electronics devices are designed for intuitive operation and with automatic protection to minimize human error.

The key point to me is that any Li charger with any automatic features must be designed so that it automatically defaults to the safest setting if there is no user intervention. This is not the case with the AF 109 if the autodetect feature can change cell count and continue into an over-voltage condition in the absence of user intervention.

Therefore, I do not see this issue going away.
Amen,
What does Astrobob have to say about this? Can something be done? Will something be done?

rc
RCBABBEL is online now Find More Posts by RCBABBEL
RCG Plus Member
Reply With Quote
Old Dec 01, 2005, 12:11 PM
know it all
Miami, Fl
Joined Nov 2003
2,949 Posts
whatever he does, I will probably buy a 2nd 109.. darn lipos take too long to recharge at the field. Maybe try 2c-3c with a fire extinguisher handy

Herm

Quote:
Originally Posted by RCBABBEL
Amen,
What does Astrobob have to say about this? Can something be done? Will something be done?

rc
hermperez is offline Find More Posts by hermperez
Reply With Quote
Old Dec 01, 2005, 01:15 PM
Flying motor mount master
fly_boy99's Avatar
San Jose, California, United States
Joined Oct 2004
8,745 Posts
In the new age of open source everything... maybe Bob can open up his source and we can program our own features!!!

EEPROM programming isn't all that hard... well I guess that easy for me to say but hey I think you will gain alot more function/features this way instead of bothering Bob whom I think doesn't want to change much after my conversations with him.

ymmv,
fb
fly_boy99 is offline Find More Posts by fly_boy99
Reply With Quote
Old Dec 01, 2005, 01:18 PM
Suspended Account
USA, FL, Apopka
Joined Dec 2000
4,072 Posts
At the heart of this deabte is weather or not a user can be trusted to make the right decision.

With the Astro 109 that is:

- check the cell count at the end of stage 1 (3 minutes)
- don't charge a charged pack

One proposed solution is to insert a new "manual confirmation" by the user. But as Dave H correctly posted:

"Requiring the user to monitor an automatic process and halt the process if it is incorrect is not intuitive for most people, as this is not a common situation with modern consumer electronics devices."

My solution is to fix the stage 1 process so that the posibility of a cell count bump is eliminated for all packs under 4S.

What about the high cell count users?????

IMHO the users of high power Lipo packs can by expected to be better informed and follow the instructions. As I have pointed out previously in many cases the charging paramaters of these packs require a cell bump.

If a better stage 1 algorithm can be developed to also handle high cell count GREAT!!!!

But the greater problem to fix first is the less informed user community that flys 2S and 3S packs and can either not follow instructions or have gotten complacent and/or lazy. This can be fixed via software changes to stage 1.

ASTROBOB........ are you listening?

Brad
bradpaul is offline Find More Posts by bradpaul
Reply With Quote  (Disabled)
Old Dec 01, 2005, 01:29 PM
Flying motor mount master
fly_boy99's Avatar
San Jose, California, United States
Joined Oct 2004
8,745 Posts
Bjpaul-

I think you may be on to something...

I don't believe I've seen a report of a lipo fire over a 5S pack???

Am I wrong?
fb
fly_boy99 is offline Find More Posts by fly_boy99
Reply With Quote
Reply


Thread Tools