AC DC Adaptor fail

[DiyNutJob]

Hit a solid brick wall, and the battery made no progress for several days. I have one more option of upping the voltage by using the high power adaptor. I have misplaced that, which forces an end to the experiment. I will resume this once the adaptor is found amongst the mountain of junk I have.

Battery state:
13.91V while connected to charger, draw from mains 1.8W, feed to battery 18V 27mA pulses at 23.27kHz. average feed voltage 2.7v
13.71V stand alone
6.39 m-Ohm internal resistance
87% wear life
406 CCA, Spec: 435 CCA, type 075
Bubbling rate closest to ear at 1 inch distance: 5 per second, subjective

 

[DiyNutJob]

I put the battery on the high power adaptor and charged at 21V, 1.8w overnight, both the CCA and IR got worst. The solid brick wall is confirmed, and nothing further could be done with the battery.

Next, I will find out where the brick wall is for the 027 battery. This time, I will use the 21V adaptor at 1.8W. Judging by the earlier small CCA and IR drop it caused, this adaptor is slightly overpowered but not by very much.

Battery state: 2023-05-16
12.85V stand alone
4.33 m-Ohm internal resistance
100% wear life
599 CCA, Spec: 550 CCA, type 027

Battery state:
12.69V stand alone
4.4 m-Ohm internal resistance
100% wear life
589 CCA, Spec: 550 CCA, type 027

 

[DiyNutJob]

The battery reached a high water mark of 604cca and bounced off and oscillated slowly in a range below that. The size of the oscillation correlated with the size of the power feed. Whether a charger is over powered can be judged using the size of the oscillation. My preference is for minimal over power. Once the charger is disconnected, and the battery is allowed to sit idle for some hours or a day, the CCA would return to the max level and then the battery self-discharges naturally and slowly after that.

I am fairly certain the battery has 600cca from the factory, but the ebay seller under-spec'ed it to cover themselves from potential complaints. So, I am re-spec'ing it as a 600cca battery myself. The battery can be considered good as new. Not bad for a workshop refurbished battery that must have been in use for some years prior. I got it for £30 a couple of years ago. 600cca is mid-range for the 027 type. The top end is 640cca. I was hoping mine was a 640, but no luck this time.

 

[DiyNutJob]

Crazy! The CCA is still rising with the battery not connected to anything for 3 days now. The battery, the charger, or whatever has gone above and beyond my expectations. This is suggestive of the need for resting/power-drawing periods for the battery within the charging process.

Battery state:
13.03V stand alone
4.18 m-Ohm internal resistance
100% wear life
620 CCA, Spec: 600 CCA (DIY re-spec'ed), type 027

 

[DiyNutJob]

The internet says temperature rise causes battery internal resistance to drop - translating to higher CCA. To rule out the temperature being a factor in the increasing CCA, I measured the 075 battery stored at the same location. The CCA of that dropped by 1 CCA since yesterday. So, the higher temperature of today is unlikely to be a significant factor.

 

[DiyNutJob]

Here's the final prototype for my charger controller. It uses a standard china 5A 4.5V-30V DC motor PWM controller that generates pulses of the input voltage and current at 20khz with variable duty cycle. For purpose of battery trickle charging, the minimum duty cycle can be used. The output to the alligator clamps are gated by a couple of 1n5391 diodes. Voltage feedback from the battery to the controller stops the controller from working properly. The diodes block this feedback. What remains missing is a couple of fuses to protect the controller from the battery, or vice versa. A fuse holder has already been attached to the red clamp. The red lead will be connected to the other end of the fuse holder.

I believe high voltages charging is beneficial. But high voltage at 14.4V or above causes loss of water. The compromise I have chosen is to pulse high voltage in very brief periods. This should reduce or eliminate water loss. I am willing to bet my batteries on this theory that I pulled out from thin air.

 

[DiyNutJob]

Today's temperature being lower, no change on the CCA is measured. The CCA has stabilised and stopped moving up. With the battery fully rested, I am putting it back on charge to see if I could improve the CCA further. It is now apparent the CCA is not final immediately after charging. The energised electrolyte negates some of CCA. Once this energy dissipates, an accurate measurement of the CCA is then possible. It seems plausible bubbling and energising of the electrolyte are related because as soon as the battery goes on high voltage charging, the CCA begins to decline. The decline is proportional to the charging power. The highest decline I have seen is about 30 CCA. After that, it bounces back up, but would never reach the starting point CCA.

Battery state:
12.98V stand alone
4.18 m-Ohm internal resistance
100% wear life
620 CCA, Spec: 600 CCA, type 027

 

[DiyNutJob]

Although the PWM controllers work, these are not ideal because I didn't know better. Next I am getting a proper controller. This will give me a mini lab-PSU and finer control over the charging power. This 3A unit can step up as well as step down the input power, giving me the ability to use all kinds of ACDC adaptors.

 

[DiyNutJob]

Further experimentation with high voltage (24v) and low average amperage (<5mA) using PWM controllers produced no gains. Nor, did it make things worst. This configuration would have charged batteries but not able to rejuvenate worn batteries.

Then the mini lab PSU arrived. It offered significantly better control over the output. But it had some limitations: for instance I could step up the 24v 1A adaptor to give me 26v 10mA. In order to get 30v, the output amps had to be 20mA.

At 30v 20mA, nothing exciting was happing to the battery being charged. At 30v 40mA, things started to move, with both the battery voltage and CCA being lifted. To replicate the solar panel that was capable of charging a battery to 16v, I may have to increase the feed amps to battery to 80mA. At 30v 40mA, there was light bubbling in the electrolyte, which I consider subjectively to be acceptable, especially when resting intervals are used during charging. 30v produced uncomfortable arc'ing that was eating away my multimeter probes and crocodile clips. The solution was to turn off the charger when connecting or disconnecting it to the battery.

Currently, using 30v 40mA, I got the 075 battery to the max 406 CCA level previously attained. It seems hopeful this could be exceeded. If not, I still have the head room of another 40mA at my disposal. A lot more mA's are available, but the electrolyte bubbling would become unacceptable.

 

[DiyNutJob]

24v stated in the previous post should have read 21v. The AC/DC adaptor I am using is rated at 21v 1A.

Charging at 30v 40mA did break through the highest CCA seen on the battery previously. But, the break through is only 2 CCA. 408 CCA is going to be the hard limit for the 075 battery. The battery came from Tayna Batteries many years ago because it was the cheapest they listed. I am fairly certain they cheated and over-spec'ed the battery at 430 CCA. The same battery being sold by anyone else is rated 400 CCA. Current Tayna's image on left; another current amzon seller's image on right:

Based on my other data, I am certain the battery is 400 CCA. Therefore, I am DIY re-spec'ing it to 400 CCA. After self-discharge, the battery sits around 10-15 CCA below spec. This is reasonable for an old battery. I consider it good as new after the trauma of being run down to 4v during the last winter.

Charging at constant 30v 80mA produced too many electrolyte bubbles, and that is not a viable solution. Using pulses of it will be fine.

 

[DiyNutJob]

I can confidently re-spec the battery CCA down because charging at 30v has confirmed rejuvenation abilities. If rejuvenation is not happening, then we are looking at a hard physical limit.

Charging at 30v 40mA lifted the peak CCA on the 027 battery from 620 to 636 CCA. Although this is only achieved after resting the battery for a day following charging. It now looks like I have a top end 027 battery on my hands. The 027's max out at 640 CCA and mine is so close. It needs a teenie, weenie push to reach the finishing line. This is where my next prototype comes in. Through pulsing, I could ramp up to 80mA or more without increasing bubbles to an unacceptable level. The chosen pulse controller is the standard 3A china PWM. It has wider low-end duty cycle range than the others. This is also the controller that will go full throttle at random moments from feedback from the battery. The feedback is prevented through use of diodes on the controller's output. I will add the diodes if I see full throttle appearing. Full throttle at 30v 80mA is still within my comfort zone, even though I won't like it.

 

[DiyNutJob]

The CCA on the 027 is still going up by itself. Last check was 3 hours ago. Now standing at 638 CCA, 12.91v, 4.07mOhm.

 

[DiyNutJob]

I have put aside the 027 and took the 19 year old 096 out of the car. The 096 started the engine with hesitation, as was expected. Stopping and starting in a shopping area was too much for it and almost stranded me. Leaving to charge by the perma-connected solar panel for 45 minutes while I shopped, gave it just enough juice to get me started for home. Obviously, that's not an acceptable experience. Rejuvenating this battery now that I have the tentative means to do it would be the priority.

After coming out of the car:
12.82V stand alone
13.62 m-Ohm internal resistance
190 CCA, Spec: 640 CCA, type 096

Currently the battery stands at 205 CCA, exceeding what was achievable in the earlier part of the thread. It is being charged at constant 32.44v and 120mA. The bubbling is low and acceptable. It's becoming apparent the bubbling level is the prime determinant for how much charging power can be used. With the other batteries, 40mA with high constant voltage is the most they can take. To play this game, you need a good ear.

 

[DiyNutJob]

Battery state:
13.52V while connected to charger, draw from mains 3.3W, feed to battery constant 32.44V 120mA
13.39V stand alone
12.46 m-Ohm internal resistance
11% wear life
208 CCA, Spec: 640 CCA, type 096
Bubbling rate closest to ear at 1 inch distance: 2 per second, subjective

 

[DiyNutJob]

Finally admitted defeat on the 19 year old battery and retired it to the recycle center. The max CCA that could be restored to it was 209. The internal resistance is 3x normal. Consequently the output power is 1/3 normal: 640 CCA (spec) / 3 (3xresistance) = 213 CCA max available. The battery is usable for the summer, but I need the space more than the battery. So, out goes the battery. Currently have 3 batteries for 2 cars. Fairly low chance both cars would be flat at the same time. So, 1 standby battery is adequate.

The electrolyte in the battery is in perfect condition. Everything is down to the internal resistance. I speculate Bat-Aid works very well (confirmed with actual experience over the years), but it permanently plugs up the battery plates resulting in the observed internal resistance. Age also causes IR. I remain convinced aging is reversible with high voltage. This belief can be tested on the remaining sealed batteries.