Author Topic: My Changing Dump Rate on Ghurd Controller  (Read 8948 times)

0 Members and 1 Guest are viewing this topic.

Offline WooferHound

  • Technowhiz
  • Global Moderator
  • Hero Member
  • ******
  • Posts: 897
  • Karma: +40/-3
  • Huntsville Alabama USA
    • My personal webpage
My Changing Dump Rate on Ghurd Controller
« on: January 24, 2012, 10:49:00 am »
Hey Ghurd
I had talked to you in E-Mail in the past about the Dump Rate changing Speeds on my Ghurd Dump controller. The dump rate will alternate between Fast & Slow in 2 second intervals. You had sent me a replacement controller and it did the same thing. Then I bought 2 more controllers and the one that I installed is also doing the same thing again. This is not a bad problem and is really only barely noticeable. I do not consider this to be a design problem or defect.

You mentioned in another post (somewhere) that the LED indicator is tied right onto the output drive circuit, but the dropping resistor for the LED should not allow it to effect normal operations. However I have 2 LED indicators tied to that output through 1000 ohm resisters. One LED is on the Dump controller itself, and the other LED is 110 feet away in the house through telephone wire. Both LED resistors are on the controller board.

Could my double LEDs be effecting the way the controller operates ?
----- W o o f e r h o u n d -----
My Renewable Energy Projects

Offline ghurd

  • Global Moderator
  • Sr. Member
  • ******
  • Posts: 442
  • Karma: +22/-0
    • GHurd Solar
Re: My Changing Dump Rate on Ghurd Controller
« Reply #1 on: January 24, 2012, 06:31:23 pm »
Double Leeds should have no effect like you are describing.
It would reduce the hysteresis a tiny bit.  It should stay stable under identical conditions.

The only thing I can think of that would cause something like that is a small load on the battery that changes every 2 seconds.
G-

Offline Rover

  • Full Member
  • ***
  • Posts: 194
  • Karma: +7/-1
Re: My Changing Dump Rate on Ghurd Controller
« Reply #2 on: January 24, 2012, 07:21:34 pm »
Probably his fancy strobe unit

sorry Woof... grin
Rover
Location: South East Virginia US

(Where did I bury that microcontroller?)

Offline WooferHound

  • Technowhiz
  • Global Moderator
  • Hero Member
  • ******
  • Posts: 897
  • Karma: +40/-3
  • Huntsville Alabama USA
    • My personal webpage
Re: My Changing Dump Rate on Ghurd Controller
« Reply #3 on: January 24, 2012, 09:32:03 pm »
Thanks Ghurd
I don't consider this to be a problem.
When it's dumping at high frequency at alternates between 50 & 60 flashes a second, the rate is too fast to see by eye but it appears to go bright-dim-bright-dim every 2 seconds, again this is barely noticeable.
Don't have a Wind Turbine to watch so I am probably paying too much attention to my solar Dump Indicators.
----- W o o f e r h o u n d -----
My Renewable Energy Projects

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: My Changing Dump Rate on Ghurd Controller
« Reply #4 on: January 24, 2012, 09:48:59 pm »
You mentioned in another post (somewhere) that the LED indicator is tied right onto the output drive circuit, but the dropping resistor for the LED should not allow it to effect normal operations. However I have 2 LED indicators tied to that output through 1000 ohm resisters. One LED is on the Dump controller itself, and the other LED is 110 feet away in the house through telephone wire. Both LED resistors are on the controller board.

Silly question: but why did you put the remote LED on the FET drive side?
 (a) Any induced noise in the long 110 foot cable will be coupled onto the FET gate, and may affect its operation
 (b) It would seem far more "sensible" to me, to put the LED in parallel with the actual dump load itself.

Option (b) overcomes the (potential) problem of (a), and also gives you positive indication that the FET is actually working too.

Just my 2c.

Offline WooferHound

  • Technowhiz
  • Global Moderator
  • Hero Member
  • ******
  • Posts: 897
  • Karma: +40/-3
  • Huntsville Alabama USA
    • My personal webpage
Re: My Changing Dump Rate on Ghurd Controller
« Reply #5 on: January 24, 2012, 09:55:53 pm »
It's Ghurds kit. I have not modified it except to put 2 LEDs instead of one, and use 1000ohm resisters instead of 470ohm.

I had never considered about induced noise possibly firing the FET. The wire is 110 feet of CAT 5 wire and the individual LEDs are on separate twisted pairs.
----- W o o f e r h o u n d -----
My Renewable Energy Projects

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: My Changing Dump Rate on Ghurd Controller
« Reply #6 on: January 24, 2012, 10:04:30 pm »
I had never considered about induced noise possibly firing the FET. The wire is 110 feet of CAT 5 wire and the individual LEDs are on separate twisted pairs.

OK, I just looked at the circuit. When you said "Drive", I mistakenly thought the LED may have been on the FET "Drive" side. It appears the standard LED is from Drain to Ground, and shouldn't be the problem I thought.

Looking at it though - the 470R resistor originally in series with the LED, probably should be 470R for *EACH* LED.

While there "shouldn't" be a problem with the way you've connected it, there is a chance that higher-frequency noise is being carried back to the controller and affecting its operation. Have you tried disconnecting the long cable and see if the issue remains? (disconnect *ALL* the long wires, not just one side of each!).

If disconnecting the wires DOES make it behave "normally", I'd reconnect them and try bypassing them with perhaps 100nF to see if that tames it down. I'm sure Ghurd will make more intelligent suggestions though.

Offline WooferHound

  • Technowhiz
  • Global Moderator
  • Hero Member
  • ******
  • Posts: 897
  • Karma: +40/-3
  • Huntsville Alabama USA
    • My personal webpage
Re: My Changing Dump Rate on Ghurd Controller
« Reply #7 on: January 24, 2012, 10:26:19 pm »
Some good suggestions
Yes I can easily disconnect the long wires as they are connected through a terminal strip. However I have had very little Sun this month and have not had much actual dumping happening. Almost made it to full charge today but rain is moving in again and won't see Sun again for 3 or 4 days.

I'm just a few miles away from Redstone Arsenal and Marshall Spaceflight Center. There is a lot of military and about 1/4 of NASA. I could be getting some interesting signals from over there.
----- W o o f e r h o u n d -----
My Renewable Energy Projects

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: My Changing Dump Rate on Ghurd Controller
« Reply #8 on: January 24, 2012, 10:32:13 pm »
I'm just a few miles away from Redstone Arsenal and Marshall Spaceflight Center. There is a lot of military and about 1/4 of NASA. I could be getting some interesting signals from over there.

Ooooohhhhh yeaaaaaahhhhh!

2 seconds. Is it a roughly equal mark/space ratio? (if it's cyclical every 2 seconds, is it 1 second on, 1 second off? Or more like a blip every 2 seconds?)  That could even be a mid-range radar sweeping past.

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: My Changing Dump Rate on Ghurd Controller
« Reply #9 on: January 24, 2012, 10:47:36 pm »
OK, I just looked at the circuit. When you said "Drive", I mistakenly thought the LED may have been on the FET "Drive" side. It appears the standard LED is from Drain to Ground, and shouldn't be the problem I thought.

Some days I'm so blind I amaze myself.

I take that back. Yes, the long wires *ARE* onto the gate of the MAIN load-switching FET, and could very easily cause the symptoms. The method of testing still stands though, as does my initial suggestion to tame it.


Offline ghurd

  • Global Moderator
  • Sr. Member
  • ******
  • Posts: 442
  • Karma: +22/-0
    • GHurd Solar
Re: My Changing Dump Rate on Ghurd Controller
« Reply #10 on: January 24, 2012, 11:28:29 pm »
The standard Yellow LED resistor is 3.3K.
The Yellow LED typically uses about 3.3ma, though it varies a bit depending on the LED and specific purpose of the controller (ie: LVD).

R4 = 130 ohm
R5 = 8.2K typical, could be vary widely depending on the application.

Never had any noise issues before. 
110' is a long ways.
I have no clue what radar would do to it.

Ross,
Is high HZ noise like that visible to the human eye?
I do not understand how it could oscillate slow enough to be detectable by eye.

Woof,
After trying it with the long wires disconnected, and if that solves the issue,
Might try connecting the remote Yellow LED in parallel with one fet's resistors, and a 100nF cap too.  It couldn't hurt.

G-

Offline rossw

  • Senior Moderator
  • Hero Member
  • *******
  • Posts: 879
  • Karma: +35/-0
  • Grumpy-old-Unix-Admin
Re: My Changing Dump Rate on Ghurd Controller
« Reply #11 on: January 24, 2012, 11:39:09 pm »
Never had any noise issues before. 
110' is a long ways.
I have no clue what radar would do to it.

Ross,
Is high HZ noise like that visible to the human eye?
I do not understand how it could oscillate slow enough to be detectable by eye.

Just hypothesising for a moment, if there is "some sort of" signal that is far too high frequency for the eye to see, it could be inducing just enough bias (remember, we've got a LED - so a diode junction), that any stray 50/60Hz mains signal could be superimposed on it as well.

That could (I suppose, theoretically), modulate the FET at 50Hz.

Offline ghurd

  • Global Moderator
  • Sr. Member
  • ******
  • Posts: 442
  • Karma: +22/-0
    • GHurd Solar
Re: My Changing Dump Rate on Ghurd Controller
« Reply #12 on: January 24, 2012, 11:51:36 pm »
(60Hz in Wooferville)

Something unusual is going on there.
It's intentionally not common to get more than maybe 10 or 15Hz, though it does "adjust itself" as required.

Unusually high frequency is almost always due to connecting the controller wires, and the load wires, to the same wires.  Resulting in oscillation.
G-

Offline WooferHound

  • Technowhiz
  • Global Moderator
  • Hero Member
  • ******
  • Posts: 897
  • Karma: +40/-3
  • Huntsville Alabama USA
    • My personal webpage
Re: My Changing Dump Rate on Ghurd Controller
« Reply #13 on: January 25, 2012, 12:14:54 am »
All versions of the Ghurd controller that I've made have been able to flash so fast as to appear to be steady on. My Hysteresis is set to .01 volt as per Ghurds original recommendation years ago. Way back when I had a single stage dump controller, I hooked some headphones up to the LED output so I could try and determine the dump frequency by listening to the clicks. I do Live Sound and I can tell you the frequency by listening. What I was hearing was a 50hz tone.

I positively have 2 separate wires working the unit, one set for Dump power and the other set for battery voltage sensing.

RossW
Well, the long LED wire is running right next to a 120v 60hz power feed. It is possible that the 50 & 60 hz frequencies are beating together to give me some 2 second variations in Dump Frequency
Thinking about that again, It happens at lower dump rates too. even 2 dump pulses every second will alternate between fast and slow every 2 seconds or so.

I will not have enough Sun to cause it to dump for about 5 days, so I will try disconnecting the long LED wires whenever that happens to try and trouble shoot this nonproblem.
----- W o o f e r h o u n d -----
My Renewable Energy Projects

Offline WooferHound

  • Technowhiz
  • Global Moderator
  • Hero Member
  • ******
  • Posts: 897
  • Karma: +40/-3
  • Huntsville Alabama USA
    • My personal webpage
- U P D A T E -
« Reply #14 on: January 30, 2012, 01:06:46 pm »
- Update -
OK the Sun has finally come out for a few days and and it took a while to charge the batteries enough to start dumping. I have been watching the 2 dump indicator LEDs and everything seems to be working perfectly. The temperature has dropped considerably and that May be effecting the operation a bit. Both dump stages are working simultaneously right now and it is very regular timing.

I have decided to rewire my Dump Controller a little bit and hook the LEDs on the long wires onto the Dump Resisters themselfs.
----- W o o f e r h o u n d -----
My Renewable Energy Projects