26984 Total Charge III voltage has me stumped

[Pirate Dave]

JohnnieB & Radioman: I sincerely appreciate your support.
More info: nothing connected to the heat sink except the diodes. Voltage going to the diodes is 91VAC (45VAC on each leg back to center tap). Still only get 42VDC from heat sink to center tap. That makes me think I have a half wave rectifier also. I measure resistance from heat sink to chassis and it's open circuit (no short). I hook up an external full wave bridge rectifier across the transformer and sure enough 82VDC (Note: I am not using the center tap). Hook things back to normal (still only 42VDC) and I'm thinking bad diode again. I remove one of the wires going to the diodes and voltage drops to 21VDC same thing on the other leg which tells each diode is providing half so it is a full wave rectifier. I don't have experience with ferro-resonant transformers, but I'm guessing Radioman may be right about the cap size. The original was bad and so rusty I couldn't tell what the value was. I read in a post that it should be about 20uF for a 36V charger so that's what I'm using. Question is: should I try more or less uF's. Thanks again for your help.

[radioman]

The old Power Wise charger I worked on had a 6uF. I replaced it with a 8uF. You may have to check around for exact value for the charger you have. JohnnieB can probably come up with a correct value. I believe your diodes are ok.

[JohnnieB]

The TC-III/IV manual I posted lists the part number as 17077-G1 and here is what that part# brings up at ShopEZGO: https://www.shopezgo.com/products/Ca...roduct_id=1551
Looks like their proofreader was asleep at the wheel. The description says "Powerwise III & IV" when it should say Total-Charge III & IV.

You ought to be able to find a 20.0 µfd capacitor for less money at your local electrical supply. Get a motor-run type instead of a motor-start type. (The run type are for continuous duty while the start type are intermittent duty.)

FWIW: The capacitance value used depends on the inductance value of the transformer, so different makes and model ferroresonant transformer type chargers use different value capacitors. 3, 6, 8 and 20 µfd are the common ones.
----------------

However, I'm still leaning towards diode issues.
You have 91VAC applied to the anodes of a full-wave rectifier, but only ~42VDC at their cathodes, yet when the same 91VAC is connected to a bridge rectifier, it yields the corrected voltage, about 82VDC.

You also said the when you connected only one diode you got ~21V from either diode.
That tells me the diodes have a lousy front to back ratio.

If I had that charger sitting on my work bench, I'd disconnect every wire attached to the heatsink and test.
If it still had low voltage, I'd remove the diodes from the heatsink and measure the DC voltage at their cathodes (Center tap to Stud).
If it wasn't about 41VDC, I'd replace the diodes with 1N1184 (or 85 or 86... The last number is the voltage 84 =100V, 85=150V,8 6=200V, 87=300V ... 90=600V)

There may be something not right about the way the heatsink is mounted in the charger, that is why I would check the diode output voltage when they were not connected to it.

[radioman]

Keep in mind that you are measuring unfiltered rectified DC. With one diode you are getting only have of the sine wave at 60 hertz. It is not true DC but rectified AC pulses and meters will not accurately measure it on the DC scale. With both diodes you are getting full wave rectification pulses at 120 hertz from both halves from the center tap. With the Bridge rectifier you also get full wave rectification pulses at 120 Hertz across the whole transformer secondary voltage. You could connect the bridge rectifier across each half of the transformer secondary and you should get the same voltage as you get with both diodes with the center tap. Since the capacitor is rusty and after spending so much time troubleshooting the thing I would replace it anyway so you will have a top notch charger when you get through in case that don't fix it. I think the capacitor is part of the problem or perhaps the problem.

[JohnnieB]

Originally Posted by radioman

Keep in mind that you are measuring unfiltered rectified DC. With one diode you are getting only have of the sine wave at 60 hertz. It is not true DC but rectified AC pulses and meters will not accurately measure it on the DC scale. With both diodes you are getting full wave rectification pulses at 120 hertz from both halves from the center tap. With the Bridge rectifier you also get full wave rectification pulses at 120 Hertz across the whole transformer secondary voltage. You could connect the bridge rectifier across each half of the transformer secondary and you should get the same voltage as you get with both diodes with the center tap. Since the capacitor is rusty and after spending so much time troubleshooting the thing I would replace it anyway so you will have a top notch charger when you get through in case that don't fix it. I think the capacitor is part of the problem or perhaps the problem.

I thought he already replaced the cap and was wondering if he had used the right value.

You may very well be right about the capacitor.

[Pirate Dave]

Since I'm using a new 20uF run cap I will try a lower value tonight. I also have an 12, 8 & 5uF, but may not be able to try all: wife has a Honey-Do list Side note: there's nothing connected to the heat sink except the diodes which I removed, cleaned and reinstalled already. Resistance from heat sink to chassis reads open circuit. I'm voting for the cap

[JohnnieB]

I might be wrong, but I think the 91VAC open circuit voltage will decrease when the lower µfd cap is installed.
Let us know what the voltage is with the 12 µfd.

If you happen to have an analog VOM lurking around there someplace, try using it instead of a DMM.
I'm pretty sure the Lester transformer/capacitor test was originally done with an analog VOM.

[radioman]

No mention of it being connected to a golf cart and failing to charge. I just assume it didn't work right and that is why it is being worked on. All the measurements will change if connected to a golf cart to filter the ripple. I am beginning to wonder if there is anything wrong with it.

[JohnnieB]

It would be nice to know the back-story.

What did the charge do, or not do, that prompted getting inside the case?

[Pirate Dave]

I apologize for not providing a complete background; I just didn’t want to make it lengthy. The charger belongs to a friend who said it would not turn on and asked if I could take a look at it. He had already taken the cover off and it has significant corrosion. Nevertheless, I started reading some threads on this site and suspected the circuit board so I disconnected it and bypassed the relay, that’s when I discovered the low voltage. I checked the cap and it was bad so I replaced it with a 20uF run cap as suggested. Then I started asking for help, which I do appreciate.

I think we're starting to make some progress as I mentioned before I have no experience with ferroresonant transformers and from what I read they seem to be sensitive to capacitor value. Last night I decided to experiment with different caps (remember the circuit board is still disconnected). I installed the caps and measured the open circuit voltage as well as voltage with a 125Ω load & 1,000uF cap across it to stabilize ripple. These are the results:

Cap DCVoc DCV with load & cap
20uF 42 47
15uF 41 46.5
12uF 41 46
10uF 41 56
6uF 77 66
4uF 71* 87
3uF * 67
*Note: With a 4uF cap the initial open circuit voltage was 0 then jumped to 71V after about 5 seconds; there was no hesitation with the load connected. With a 3uF cap the output was unstable even with a load connected.

I would like your thoughts on which cap I should use before connecting the circuit board and trying it on the cart? Any other suggestions are greatly appreciated. Many thanks.