MOSFET voltage regulator/rectifiers

[User Bp]

The RR on my '01 sv650s seems to be on the way out.
Shindengen's fh008 seems to be a popular replacement
and are said to run much cooler that the SCR style.

New units don't seem available, but madhornets.com
has an fh002 which seems promising. However, I can't
find specs. It's rumored to be a 25 amp unit, which
would work.

If anybody's been through this drill before I'd be
grateful for your insights.

Thanks for reading,

bob prohaska

[Mark Olson]

Post by User Bp
The RR on my '01 sv650s seems to be on the way out.
Shindengen's fh008 seems to be a popular replacement
and are said to run much cooler that the SCR style.
New units don't seem available, but madhornets.com
has an fh002 which seems promising. However, I can't
find specs. It's rumored to be a 25 amp unit, which
would work.
If anybody's been through this drill before I'd be
grateful for your insights.

The components of most any three-wire permanent magnet motorcycle
alternator are pretty much interchangeable, electrically. Do the
connectors on the FH002 match the ones in the SV wiring harness? If
so, it should plug and play, otherwise you'll need to get out the
crimping tool and heat shrink.

The madhornets.com website says the FH002 fits Honda CBR600,
CBR900/954/929, CBR1100XX. Unless those all have significantly weaker
alternators than the SV I can't see how it wouldn't work.

25A should be fine, I doubt the stock SV650 stator puts out much more
than 20A.

PS it's still probably a shunt regulator (they make MOSFET series
style, too) but the voltage drop across the FETs is lower than an SCR
so it dissipates less power when shorting the windings for regulation.

https://www.shindengen.co.jp/product_e/electro/reg.html

[User Bp]

Post by Mark Olson
The components of most any three-wire permanent magnet motorcycle
alternator are pretty much interchangeable, electrically. Do the
connectors on the FH002 match the ones in the SV wiring harness? If
so, it should plug and play, otherwise you'll need to get out the
crimping tool and heat shrink.

No expectation of plug and p[l,r]ay, I have a power take off splice
that will require re-terminating the RR anyway. The more I think of
it the better the experiment seems.

Post by Mark Olson
The madhornets.com website says the FH002 fits Honda CBR600,
CBR900/954/929, CBR1100XX. Unless those all have significantly weaker
alternators than the SV I can't see how it wouldn't work.

That's consistent with my thinking.

Post by Mark Olson
25A should be fine, I doubt the stock SV650 stator puts out much more
than 20A.

The manual rates the generator at 300 watts, but they don't state
a nominal voltage. It's implausible the current is more than 25 amps.

Post by Mark Olson
PS it's still probably a shunt regulator (they make MOSFET series
style, too) but the voltage drop across the FETs is lower than an SCR
so it dissipates less power when shorting the windings for regulation.

I'm very unclear on the shunt-vs-series tradeoff. Shunt regulation maximizes
copper loss, but series regulation takes the core material through its entire
magnetization curve, maximizing iron loss. The voltage isn't a problem, but
hysteresis in the iron might be, depending on how the designer balanced the
losses. With enough copper shunt might be more efficient, an excess of iron
would favor series regulation. Most likely the optimization was for cost,
it's not obvious where that puts matters.

Post by Mark Olson
https://www.shindengen.co.jp/product_e/electro/reg.html

Thanks for reading,

bob prohaska

[Mark Olson]

Post by User Bp
The manual rates the generator at 300 watts, but they don't state
a nominal voltage. It's implausible the current is more than 25 amps.

The wattage figures I have seen usually at calculated at 14V or so,
which is a reasonable battery charging voltage, so about 21A max.

Post by User Bp

Post by Mark Olson
PS it's still probably a shunt regulator (they make MOSFET series
style, too) but the voltage drop across the FETs is lower than an SCR
so it dissipates less power when shorting the windings for regulation.

I'm very unclear on the shunt-vs-series tradeoff. Shunt regulation maximizes
copper loss, but series regulation takes the core material through its entire
magnetization curve, maximizing iron loss. The voltage isn't a problem, but
hysteresis in the iron might be, depending on how the designer balanced the
losses. With enough copper shunt might be more efficient, an excess of iron
would favor series regulation. Most likely the optimization was for cost,
it's not obvious where that puts matters.

Magnetics were never my strong suit ("I understand some of those words")
and I opted to skip all the power and rotating machines stuff... but my
guess is that even using an ideal switch with zero resistance, the stator
isn't going to overheat, compared to how it would function with the stock
SCR-style shunt regulator.

[User Bp]

Post by Mark Olson
guess is that even using an ideal switch with zero resistance, the stator
isn't going to overheat, compared to how it would function with the stock
SCR-style shunt regulator.

Agreed entirely. The order to Mad Hornets is in, let's see what happens.

Thanks for reading!

bob prohaska

[User Bp]

Post by User Bp
Agreed entirely. The order to Mad Hornets is in, let's see what happens.

In case anybody's curious, the Mad Hornets order arrived in
six days (from China) by USPS. To that extent I'm impressed.

The regulator delivered bears a "YHC" logo in addition to the
Shindengen-like part number FH002 4.0 251 and the Shindengen
trademark (two inverse-series diodes in a circle).

Hooked up to my VFR800 it worked reasonably. It got warmish, but
certainly no worse than the OEM RR. The worrisome thing was that
most of the heat seemed to emanate from a very small area under
the part number markings. In that small area it was too hot to
touch comfortably after a few minutes' idling. Regulating voltage
was around 13.7 on a Shorai LiPO4 battery, rising to just shy of
14 when revved. That's low, but it usually takes a while to peak
and given the heat I didn't want to run it that long.

Granted, the test is a little unfair. The FH002 is said to be a
25 amp unit, the VFR alternator is a 34 amp device. It seems worth
intalling in the SVS with a 24 amp alternator to see if the heating
is more manageable.

Whether this is really a Shindengen FH002 MOSFET regulator, a licensed
copy, an unlicensed copy or a conventional SCR regulator sporting a
more attractive part number remains unclear.

Thanks for reading,

bob prohaska