DC Voltage Drops While Running

[blaat!]

Sorry for another long delay. Fall trail riding in New England is currently the priority! The bike is now in the basement for the winter…unless we get some warm days. Word is that there will be another MotoGiro in the Pocono Mountains of Pennsylvania mid May. If I don’t have a conflict, I’ll be targeting that. Last one was such a blast!!!

mrkprsn - I think the voltage came up as a result of a combination of disconnecting the (OEM) ground in the stator, using the Electrex World reg/rec, and running with a battery (a.k.a. capacitor) which is required for that reg/rec. I’m not sure that I previously had all three in place at the same time during testing or riding. I’m still not convinced all is right, but I’ve never measured that much voltage before so moving in the right direction.

Nigel - thanks for the idea. I’ll test it out depending on how the voltage looks going forward.

Nick - my bike seems to be putting out the right amount of amps from the generator so I think the flywheel magnetism is okay!?! Would love to check it though. Unfortunately a lot of the tools and knowledge is disappearing

Max - I won’t rule out an overfilled battery if it stops leaking before running low on liquid then I will know.

Thanks all. I’ll update more as it comes. Once this is confirmed sorted I’ll try to make a summary post to save everyone from reading this whole thread!

[blaat!]

Just back from the MotoGiro in the Pocono Mountains of Pennsylvania, USA. Around 450 kilometers, 110 bikes, all pre 1969, and under 350cc (or something like that).

Giro.jpg

I fried two more batteries over the weekend. Both read 4.5V now and neither will charge with trickle or straight connected to a DC power supply. Dropped a cell on each?

This is the graveyard of batteries at this point, minus one more wet acid I already recycled. Two different regulators plus one more that MrkPrsn gave me at the Giro (great to meet another Motoscrub in person!) as well as the original regulator.

batts-n-regs.jpg

I had onboard voltmeters all weekend. The one at the battery usually hovered between 7.2V and 7.4V (maybe a touch higher?) and the one at the coil was 6.0V-6.4V

There was one point that I pulled over on the side of the road for something else, and the bike sputtered out when I let the RPMs drop. Both onboard voltmeters read zero with everything connected and on. I disconnected the battery and measured DC voltage with my handheld meter and it was around 5V, but dropped to nearly zero when I put the battery contacts back on. I kicked the starter lever to see if the points open/closed made any difference, but no. Disconnected the battery to prevent more damage and looked for shorted wires. Took the headlight apart, tightend all the small flathead contacts inside, and put it back together. The battery read 5V with everything connected again so I bump-started it downhill in first gear to get the RPMs high enough and it eventually caught so I could finish that leg of the race. Don't know if the headlight contacts made a difference, or if something was overheated and shorting (such as the regulator or coil) and cooled while I was working?

I suspect I have too much juice now and I'm over-charging the battery. Eventually I lose a cell in the battery and bike won't idle/run at low RPMs. Then I can't get it started again if I shut it off. Mid 7Vs seems like a lot during two full days of riding.

Anyway, I think the thing I was missing during most of this thread was that the extra ground on the stator needed to be undone AND the regulator needs the battery in place to function. At many points I was trying to measure DC voltage coming from the regulator and several people have told me that the battery needs to be in place for the regulator to function correctly.

Thoughts?

[blaat!]

Also, BTW, want to get thoughts on using a capacitor instead of a battery, and if it would be more or less reliable. I believe some of the old Triumphs had a capacitor to help at starting, Lacey sells a 12V version as a battery substitute, and 12V cap battery eliminator kits are available on eBay. My thinking is that it acts to buffer the volatile voltage just like a battery?

[Jordan]

Yeah, something's wrong with your wiring.

Regarding going batteryless - what a great idea.
I have done it many times, including on my current single.
Even my Suzuki T500 would start and run, just by replacing the battery with a capacitor.
But to do it properly you should replace the regulator with a high-power type.
A Lucas ZD715 zener diode works well. An alternative is LU49345. They are 12V types.
You could use a positive or negative earth type, as there's no battery anyway.
The capacitor should be at least twice the voltage - 25V or more, and used with a shockproof mounting.
There is one possible problem, with a solution:
Sometimes when at idle, turning on the lights or horn can stop the engine.
Some wiring surgery at the alternator solves that. Ask for details.

[insanity]

Using a digital or analog multumeter - I ask because I got errant readings on a DVM on a Triumph ?

[blaat!]

Ok Jordan. If I get that far and it stalls at idle I’ll inquire more.

Insanity- it’s analog (love analog gauges!) but was a little out of calibration. I just synced it up with my power supply at different voltages and it all good now.

[Jordan]

I forgot to mention the capacitance of the capacitor.
I usually use 10,000 microfarads or up.
The original Lucas one was reportedly just 5,000 uF.

[blaat!]

I ordered a 5pk of cheap 16V 10k uF caps so I could experiment. I used Wago Lever Lock terminals so I could add/remove caps on the fly… and it worked perfectly!

It starts and runs with 1 through 5 caps hooked up in parallel. All five (or 50k uF) may be the limit… it seems to just barely catch and start on the first kick. I think it needs to charge the caps first and then the spark catches barely in time to start. I need to play around more, but ultimately I will order a couple quality brand capacitors once I know the optimal capacitance.

I went for a nice 20 mile ride with the capacitors and it ran great!

With the battery out of the picture it simplified things and helped me come back to the original title of this posting.

It was a warm day and after the bike heated up pretty well I noticed the voltage dropped below 6V.

The caps are less of a buffer than a battery so I noticed it right away. When I got back home I couldn’t get it started again.

I’m realizing it is a most likely heat issue with one of the components.

I swapped out multiple regulators with no change. The coil is a brand new Lucas so unlikely to be that.

I let it sit overnight to cool down and it started easily and was putting out about 7.8V again.

My suspicion now is that something is going on with the stator when the engine heats up. I tested the resistance of the stator across the two yellow wires:

Cold = 1.1 ohms then drops to 0.7 over about 10 seconds

Took if for another 20 mile ride, voltage dropped again, and wouldn’t start again after I shut it off. Tested the stator resistance again:

Hot = 1.7 ohms then drops to 1.1 over about 10 seconds

Any thoughts on what is going on?

Before I tested the resistance I was thinking the stator windings were expanding from the heat, and the old magnet wire coating was allowing crosstalk between the windings and producing less energy. But if that was the case I would expect the resistance to go down not up. Why would the resistance increase? And is whatever is causing the increased resistance also causing less power output?

Thanks for your input as always!!!

[Jordan]

Do you have a battery as well as the capacitors?
What is the nominal system voltage? Regulator and HT coil voltage?
Do you think you can get away long term with a regulator that is not really meant for use in a battery-less system?
I wouldn't be confident, having destroyed some "standard" regs that way.
Rather than having all your caps in parallel, try serial/parallel, with 4 caps. 50mF seems a lot, and serial connection increases voltage rating, which could improve the life.

[blaat!]

Hi Jordan,
No battery, just capacitors.

7.8V at the batt and I would say about 7.3 at the coil when I first start it. Both drop below 6V after it heats up.

No idea if it will burn out the regulator long term but I have three different regulators now so willing to risk it

My understanding is that you add the capacitance when in parallel (10K + 10k = 20k uF and the max voltage remains at 16V) but you halve the cap value and double the voltage in series (10k/2 = 5k uF but the max voltage doubles to 32V). I know it is a little more complicated than that, but 16V max is plenty for a 6V system and summing the capacitance is convenient for testing different amounts of capacitance.

The capacitors have been an insightful testing tool eliminating the buffer of the battery power storage from the equation.

If I can sort out the voltage drop when the bike heats up (suspecting the stator) I may move back to a battery again (unless the caps prove to be reliable and more maintenance free).

I believe that after the bike heats up, it is putting out too little voltage and eventually causing each battery to sulfate on longer rides. Because a battery has significant energy storage capacity, I wasn't able to see the effect right away. This would explain everything and make sense as a root cause. I'm planning to test the stator output (voltage and amperage) to confirm beyond the cold vs hot change in stator resistance values. The setup to do this is a little more complicated with a ride to heat it up in between so just need to find a few hours.

Anyone have any logical explanation for the increased resistance/lower output from the stator when it heats up?