Topic: The $6 Charging System Output Improvement Mod

[Erny]

My rectifier was still working till Nov 2020 (my last ride last season)
Bike now waits new cam and to be put back on the road, expected mid April. I expect no issue wirh rectifier

[dave500]

#$%* for all this #$%*,my used rectifier and bosch unit is solid but more than six bucks?

[caluser2000]

So as we all know, the charging system on these bikes is really marginal, especially at low RPM. I've seen a lot of talk about swapping in CB750A stators, going to permanent magnet alternators and all sorts of other stuff. Basically taking the brute-force approach to getting more power. But I personally thought there might be room for improvements in the charging system's efficiency, rather than just throwing more alternator at the issue.

The first thing that came to mind is the rectifier. Our bikes use a 3-phase bridge-rectifier to convert the alternator's electricity into DC. It does this by using 3 sets of diodes arranged like this:


The problem with this, is that standard silicon diodes have a weird characteristic voltage drop across them. For a standard diode, this voltage drop is 0.6-0.7 volts. So that means that while the amount of current through the diode can vary a lot, it's still going to have a 0.7v drop across it almost all the time. This becomes even more of an issue with a bridge-rectifier setup. For current to flow through a bridge rectifier, it most flow through two diodes, and the voltage drop of each diode stacks with each other. So a rectifier is going to drop 1.4v no matter what. That means before the alternators on our bikes can even start supplying power, they need to be generating 1.4v MORE than the battery voltage at minimum.

Since the diodes are shedding voltage, this means there is also a power loss through them that is directly proportional to the current flowing. Our bikes are quoted as having a 225w charging system, which at 14v means 16A of current. At 16A, a standard silicon rectifier is burning away over 22W of power as heat. That's a full 10% of the alternator's output!

There is a solution however (well two solutions, but more on that later): The Schottky Diode. This is a special type of diode with an extremely low voltage drop. Typically 1/3 - 1/2 that of a typical silicon diode. This means in low-voltage situations, you get a lot more of the power where you need it, and less wasted as heat: And as luck would have it, someone makes a fully integrated 3-phase Schottky Bridge Rectifier: The IXYS FUS-0045B - https://ixapps.ixys.com/DataSheet/14b6ae00-24ef-4836-9531-dd26e2e1ed59.pdf . The best part, is that this little chip only costs $6. So for $6, I was willing to take a gamble and ordered one.

Step 1 was to solder and heatshrink some wires to it. 3 wires for the incoming AC, two for the outgoing + and - leads:


Next, I rummaged through my junk bin and found a heatsink to attach the rectifier to (even though it wastes less power, it still require a heatsink to stay cool), and crimped some terminals onto the wires.


Finally, I installed the whole assembly in the bike. Now this is just a "prototype", so if the rectifier proves itself, the little clip attaching it to the heatsink will be replaced with a proper screw-retained bracket, and the spade terminals will be replaced by a proper plug.


So then I put an ammeter inline with the connection between the main harness and the battery so I could see exactly what was going on. The results speak for themselves:
Before:
1200 RPM: Battery discharging
2000 RPM: 3.2A battery charge current

After:
1200 RPM: 0.01A Battery charge current (this means the charging system is matching the bike's electrical demands exactly)
2000 RPM: 4.8 A battery charge current.

So as we can see, this $6 part is giving me an additional 1.6A of charging current at 2000 RPM. That's 20+ watts of additional power! I'm going to put some miles on this rectifier to make sure it holds up, but this is looking like a huge improvement in low RPM charging system performance for just $6 and about 1 hour worth of time to put everything together.

This place has the rectifiers in stock for about $5.50: https://www.futureelectronics.com/p/2009745?utm_source=eciaauthorized&utm_medium=CPCBuy&utm_campaign=ECIA_Ext_Refferals

I'm also looking into making another rectifier using technology called an "ideal diode bridge". This converts 3-phase AC to DC with almost 0 losses, but it requires about $35 in components and for me to make a circuit board. But with how much of an improvement just the schottky rectifier made, I think it may be worth investigating.

An excellent solution. That along with replacement leds for the bulbs would be the bees knees.

[Deltarider]

An excellent solution. That along with replacement leds for the bulbs would be the bees knees.

I've read this thread with interest and at first I thought I'd try it out. But the thing is: I have no charging problem, so why would I fix a problem that doesn't exist. I've gone through the riders reports by CB500/550 owners who together in miles went around the globe five times. Guess what: no one complained about a 'marginal charging system', not one. In this forum there seems to be a charging problem however. In absence of complaints in other fora like in Europe, I have come to the conclusion it must have to do with a problem in the US. What immediately came to mind is the dreaded headlamp-always-on wiring on the -A models and extra running lights. So I have another solution: make the headlight switchable. The (key) ignition switch will love you for that. Use the bike for what it was intended and you'll have no problem, let alone that you have to fiddle with inferior Leds which lead to fitting extra parts like diodes and stuff. Cost of my solution, in labor and in money, is practically zero. And for those that cannot withstand the desire to do extra not needed 'repairs', fit a M-unit. That will you keep you busy no end. Guaranteed.

[caluser2000]

An excellent solution. That along with replacement leds for the bulbs would be the bees knees.

I've read this thread with interest and at first I thought iI'd try it out. But the thing is: I have no charging problem, so why would I fix a problem that doesn't exist. I've gone through the riders reports by CB500/550 owners who together in miles went around the globe five times. Guess what: no one complained about a marginal charging system, not one. In this forum there seems to be a charging problem however. In absence of complaints in other fora in Europe, I have come to the conclusion it must have to do with a problem in the US. What immediately came to mind is the dreaded headlamp-always-on wiring on -A models, extra running lights. I have another solution: make the headlight switchable. The (key) ignition switch will love you for that. Use the bike for what it was intended and you'll have no problem let alone that you have to fiddle with inferior Leds which bring extra parts like diodes and stuff. Cost of my solution is practically zero. And for those that cannot withstand the desire to do extra not needed 'repairs', fit a Motogadget. That will you keep you busy. Garanteed.

I fitted a 3 LED driving light under the head light,  mainly for around town, connected to the park circuit. . But even on the open road other  traffic see it miles away. I've had quite a few divers pull over to let me pass.

I have a Halogen H4 lens from a CBX 250 that replaces the original sealed beam unit, which were pathetic to say the least, for night riding. Eventually the majority of the bulbs on the CB550F will be replaced by LEDs.

I will replace/upgrade as and when I see fit.  I weighed my bike today and it was 187kg wet, 19kg lighter than the stock 206kg wet.

Yes there have been a ton of SOHC riders and kms clocked up. What others do to there's I like to hear about certainly and help out where possible. But what really matters to me is what I do to my machine. I know every, nut, bolt, wiring/connector, components, etc of the old girl. Got it at 76,000kms up it now has over  164,000kms up. She is still running points and fires up fine first pop most of the time with fresh fuel in the carbs and charged battery. Alsdo running ball bearing in the steering head and handling is as tight as it was when released from the factory.

[Maraakate]

I was thinking about doing this mod soon as I was placing orders to Digikey quite frequently but the part listed by the OP is now NLA/Discontinued.  There is not much available on Digikey for 3P Bridge Rectifiers that are actually in stock.  I see this and it may be appropriate, but unsure: https://www.digikey.com/en/products/detail/micro-commercial-co/3GBJ3516-BP/7681643

Not sure if OP is still around and can recommend if this would do the same job, just overpowered for the application?

[PeWe]

I found by searching FUS45-0045B

https://www.micro-semiconductor.com/products/IXYS-Corporation/FUS45-0045B

Do not forget the diodes that protect total melt down when outgoing power is disconnected. Look earlier in this thread.
Plug might disconnect by accident or mistake.

I have not built mine yet. Still no connector.
I'll keep the stock, possible to just swap connector if it should fail.

[Maraakate]

I had an EE friend check it out for me and you could use this: https://www.digikey.com/en/products/detail/vishay-general-semiconductor-diodes-division/VS-36MT10/357898

It's now a $16 charging mod if you get that, but it's got a mounting hole in the center and it's overpowered enough for the application that the aluminum backplate on it is probably good enough to not require any heatsink.  Will order this sometime over the next couple of weeks and report back.

Yes, have some power diodes will add that in circuit as well.

[Korven]

I had an EE friend check it out for me and you could use this: https://www.digikey.com/en/products/detail/vishay-general-semiconductor-diodes-division/VS-36MT10/357898

It's now a $16 charging mod if you get that, but it's got a mounting hole in the center and it's overpowered enough for the application that the aluminum backplate on it is probably good enough to not require any heatsink.  Will order this sometime over the next couple of weeks and report back.

Yes, have some power diodes will add that in circuit as well.

That rectifier is made for mains voltages 230/110V and low current draw. at 10 Amps the drop is 1V, so not that much better tbh.

[Maraakate]

Ah, OK.  Nvm.  Guess I'll have to search elsewhere.

[PeWe]

Do not forget the clamping diode. Not any "power" diode.
From earlier part of this thread:

"... you could damage the rectifier with a momentary opening of the charging circuit. But there is a simple and cheap fix:

https://www.digikey.com/en/products/detail/littelfuse-inc/TP5KP18A/6174237


Solder that across the + and - outputs of the rectifier, and it will clamp the output at 20v maximum....."

Diode data from above link
Manufacturer Product Number: TP5KP18A

Description: TVS DIODE 18VWM 29.2VC P600

[maxheadflow]

Do not forget the clamping diode. Not any "power" diode.
From earlier part of this thread:

"... you could damage the rectifier with a momentary opening of the charging circuit. But there is a simple and cheap fix:

https://www.digikey.com/en/products/detail/littelfuse-inc/TP5KP18A/6174237


Solder that across the + and - outputs of the rectifier, and it will clamp the output at 20v maximum....."

Diode data from above link
Manufacturer Product Number: TP5KP18A

Description: TVS DIODE 18VWM 29.2VC P600

One of the things being missed is that this device can only dissipate 8 watts.  If you loose the connection, it will die..

[PeWe]

I finally built one.
My K6 got it.
I'll do another one for my K2.

I have had thoughts about it during low rpm rides in legal speeds, around 3000-4000 rpm.
Tested in garage.
Soon a ride to check temp on cooler and what the voltage meter between gauges will show.

http://forums.sohc4.net/index.php/topic,180468.msg2217339.html#msg2217339

Update:
I added a hanging mini fuse (15A) on the + wire. If + leg will snap off on rectifier or diode get a short, a meltdown will occur between rectifier and battery.

[HondaMan]

I think I will build up some of these to test on the 750 and 550 I have here. My old ones are 50+ years old now! The old selenium type rectifiers degrade over time, losing PIV rating each time they get overheated. The SOHC4 designs had silicon rectifiers to forestall this degradation, but a measurement while running will tell the story. If it works out OK, maybe I'll make them into drop-in parts?

[MauiK3]

I'd try one!!

[PeWe]

It really is an upgrade.
I have a voltage meter between the gauges so I see the voltage levels and how it change.

It reach max quicker now than before. I had to adjust the regulator which was adjusted too high.
Crusing around at 3000-4000rpm on 5th gear, legal speeds.  Sometimes 3500rpm for a longer period of time.
The Schottky  rectifier is a good improvement.

The alu plate was warm after around 90km ride today, ca 1 hour. Gas refill.
Continued to ride another 80km, it did not become warmer.

Only warm for finger to touch, not hot. 
Ambient temp was around 24C during the ride.

[tool14]

i'd try one from hondaman.

[Tim2005]

Great. Did you buy direct from https://www.micro-semiconductor.com/products/IXYS-Corporation/FUS45-0045B ? How were they for shipping?

[PeWe]

Great. Did you buy direct from https://www.micro-semiconductor.com/products/IXYS-Corporation/FUS45-0045B ? How were they for shipping?

I bought mine from eBay 2 years ago. That seller has no rectifiers today.
The last link what I found when searching.
The clamp diode however still available from same seller.

Better to find where shipping match your location.

[Deltarider]

It reach max quicker now than before. I had to adjust the regulator which was adjusted too high.

If you suggest causality, I don't quite see how...
Unless you had (re)adjusted the regulator before and/or unless you - when you were at it this time - have cleaned contacts in the process.

[Gurp]

Great concept.
Good logic behind it. Shame American Locomotive hasn't been on in so long.

[PeWe]

It reach max quicker now than before. I had to adjust the regulator which was adjusted too high.

If you suggest causality, I don't quite see how...
Unless you had (re)adjusted the regulator before and/or unless - when you were at it this time - you have cleaned contacts in the process.

Charging works better. It is always reasons for higher load during braking, using blinkers, low rpm ride so it did not reach full for long.
When it charge better from lower rpm, it will refill battery quicker.

Another thing if I should have cruised on highways for a longer period of time. The long lasting higher levels should have become an issue.

When full voltage is reached, it will decrease after city riding, using brake light and blinkers.
It has usually taken around 20 minutes +5000 rpm riding to refill.

[grcamna2]

It reach max quicker now than before. I had to adjust the regulator which was adjusted too high.

If you suggest causality, I don't quite see how...
Unless you had (re)adjusted the regulator before and/or unless - when you were at it this time - you have cleaned contacts in the process.

Charging works better. It is always reasons for higher load during braking, using blinkers, low rpm ride so it did not reach full for long.
When it charge better from lower rpm, it will refill battery quicker.

Another thing if I should have cruised on highways for a longer period of time. The long lasting higher levels should have become an issue.

When full voltage is reached, it will decrease after city riding, using brake light and blinkers.
It has usually taken around 20 minutes +5000 rpm riding to refill.

'around 20 minutes at +5000 rpm to refill'  Was that before the schottky diode? or after ?

[Deltarider]

It reach max quicker now than before. I had to adjust the regulator which was adjusted too high.

If you suggest causality, I don't quite see how...
Unless you had (re)adjusted the regulator before and/or unless - when you were at it this time - you have cleaned contacts in the process.

Charging works better. It is always reasons for higher load during braking, using blinkers, low rpm ride so it did not reach full for long.
When it charge better from lower rpm, it will refill battery quicker.

That may well be possible, but I still don't see, why you had to (re)adjust the regulator. The way I see it, the regulator is an adjuster in itself, be it within a certain band. So, if your regulator before was set to switch to 'low mode' at say 14,5 V, what difference does a possibly better performing rectifier make? The regulator will still switch at 14,5 V, be it a bit sooner than before.

[PeWe]

It could reach 14.9V (since regulator was not correctly adjusted) which became very visible with the Schottky rectifier that charged battery to full quicker since it charged better below 4000rpm.
A normal ride has some additional loads that keep level down while riding below 4000 rpm.
If you constantly let it rev 5000-6000 rpm, another thing. The mod is an improvement for lower rpm rides.

This is exact as the measurements the starter of this thread wrote in the beginning of this thread.

Max voltage not reached when batttery is not fully charged.
Try it and see, if you have a voltmeter connected to look at while riding.