Topic: CB750K5 Starter Safety Systems

[mystic_1]

In another thread we started discussing the K5's Starter Motor Safety Unit.  (SMSU)

In the course of doing so, I decided to play with a few diagrams.


First I took a K5 wiring Diagram...



...and removed everything except for the starter safety stuff.



Next, I took the schematic of the SMSU from the CB750 69-78 manual scanned by Honda4Fun.  In the other thread we noticed several problems with this diagram.



So, I reworked it a few times, trying to move things around while keeping the electrical paths effectively the same.








Finally, I overlayed the appropriate wire color codes from the original wiring diagram.


I think this last drawing is pretty close, all of the wiring paths correspond to the way things are physically connected, the neutral light has been moved outside the SMSU bounding box, and I'm hoping I didnt screw something up along the way 

It's still a bit hinkey because of that grounded diode in the middle of the diagram.  Also there's an ambiguous junction just right of center at the resistor, the original drawing has paths crossing there but does not show either an explicit jump or junction mark.  So, anyone care to guess which it should be there?

As I said in the other thread, I just thought this would be a fun little exercise.

mystic_1

[mystic_1]

Here's a post that Grinnin made in the other thread where he describes a functional alternative to the original SMSU:

The plug&play replacement is easier than the diagram.  These 2 diodes should do everything the SMSU does.

The two diodes here are both rated for 6 amps.  One of them connects Dark Green w/Red to Light Green w/Red.  Silver band at the Light Green side.  This is the diode that provides a circuit from the starter button to the neutral switch without turning on the neutral light every time you pull the clutch handle.

The other diode connects Yellow w/Red to Black.  Silver band at the Black.  This diode works when you release the starter button and prevents the solenoid's collapsing magnetic field from creating a high-voltage spike that can burn switches.





(And yes, I'll crimp on real spades although I kinda want to put it on a full 6-circuit plug.)

I should be able to test this sometime this weekend even with the engine out.  There ARE a couple of differences.  The SMSU reduces the amount of current through the neutral switch.  The diode will put the full 3.3 to 4.2 amps through the neutral switch.  The starter button already carries the full current.  Honda uses the same part number for the neutral switch in '78 when they used a diode instead of the SMSU, so the neutral switch should be fine.

[mystic_1]

Took this diagram through a few more iterations.  Made the assumption that the "mystery point" mentioned above was NO CONNECTION.

[mystic_1]

Change History:



 

[Grinnin]

My measurements show that there is not a diode connecting Green to Black; that's the diode above the transistors.  I cannot tell EVERYTHING that's happening inside my still-sealed box, but where there's a simple diode or diode plus resistor or other simple connection, it shows up clearly with a multimeter.

It looks to me like the most accurate part of the original diagram is the diode below the starter solenoid.  A "flyback" diode is common on coils or motors that will shut off suddenly.

That means that I SHOULD be able to detect the diode to the left of the misplaced lightbulb but I cannot actually detect it.

ORIGINAL:



THANKS for taking the time to draw that up.  I'll post more if I think it helps.

[Grinnin]

I tested these this morning.  They work.  I.e. the starter solenoid responds to the starter button only if the clutch lever is pulled in or the transmission is in neutral.

[mystic_1]

My measurements show that there is not a diode connecting Green to Black; that's the diode above the transistors.  I cannot tell EVERYTHING that's happening inside my still-sealed box, but where there's a simple diode or diode plus resistor or other simple connection, it shows up clearly with a multimeter.

Yeah if you watch the progression of the diagrams, that diode you mention is the one that shows on the original diagram, just left of center, that runs straight to ground.  Since Green also connected to ground, that's where it landed when I shuffled components around.

All of this, of course, is just theoretical.  We don't know if the original diagram is correct in any way.  I'm pretty confident in my color identifications because the connection paths are now consistent with the full bike wiring diagram.  However the circuit diagram may or may not bear any resemblance to reality.

What we really need to do is to cut one open.

Anyone got a junk Starter Motor Safety Unit laying around that they don't mind destroying?



Grinnin, it would be neat to come up with an equivalent schematic that shows your two-diode replacement for this.

I'll post more if I think it helps.

Please do!

mystic_1

[mystic_1]

The two diodes here are both rated for 6 amps.  One of them connects Dark Green w/Red to Light Green w/Red.  Silver band at the Light Green side.  This is the diode that provides a circuit from the starter button to the neutral switch without turning on the neutral light every time you pull the clutch handle.

The other diode connects Yellow w/Red to Black.  Silver band at the Black.  This diode works when you release the starter button and prevents the solenoid's collapsing magnetic field from creating a high-voltage spike that can burn switches.

Like this?

[Patrick]

I think you have solved the mystery about why Honda reversed the polarity of the starter circuit in the later models. The elegance of a single diode replacing all of that....

[Grinnin]

Like this?

Yes, like that.

It's small enough to fit into any full-bike wiring diagram I've seen.

[Grinnin]

EDIT:  WHile my SMSU worked when my bike was last together, I believe I have a diode that failed open -- the flyback diode for the current through the SMSU. 

The last version from mystic_1 is almost what I read with a meter.

list]

• Yellow/red is anode (pos) and Black cathode (neg) of a diode.
• Dark Green/red is anode and Light Green/red is cathode of a diode.
• Black is the emitter of a PNP transistor and Light Green/red is base.  (This half a transistor looks electrically like a diode .)

In other news, lucky is emailing some pics of the guts of a safety unit.

[mystic_1]

To thicken the plot, I found the following post from May 2013:

Hi guys!
I have tried to clear this mysterious circuit



and make a scheme of all connexions and components.

I have the values of diodes, resistors and transistors and will try to weld it soon

He never did post back with component values

mystic_1

[mystic_1]

Relabeled my latest version to include component labels corresponding to Delamare's

[mystic_1]

And another relevant post from 2011

Makes me think some electronic genius wasted a lot of time making this"safety unit"
And now they just use a clutch switch diode with clutch switch lever switch? I wish I understood all of it better.

[mystic_1]

OK....
I wanted to find out what was in the starter motor safety unit and why all of the used ones had cracks and shrinkage on the outside case.
So I took one of the ones I had and decided to sacrifice it.
First I tried using a small exacto saw and started removing the outside case.
You can see in the photo the typical cracks and shrinkage. I wanted to know why.
I Started to remove the case and you can see the line between the case and the potting material.
I removed the potting material as carefully as possible with a small hammer and chisel.
In one photo you can see the large aluminum heat sink that caused all of the shrinking and cracking of the outside case.
I did accidentally damage one component during the removal of the potting compound.

I am wondering why this component cannot be reproduced since it seems like a simple construction. I have looked on the internet and cannot see to find any for sale.
I did see one Facebook account holder that was restoring a Honda CB750 and he had a new one that had a light grey case. He did not say where he got it.
I would ask but I do not want to become a Facebook member now that they share info with all the other social networking sites.

I do not know enough about electronics to make a new unit. I cannot identify some of the parts on the circuit board.

[mystic_1]

NOW we're talking...

From the Honda Update '74 dealer info.
Now anyone can build their own!

[mystic_1]

Here's a neat bypass suggestion from Hondaman:

I'd quote you the page number if I had one of my books handy, but... 

Find the Dark-Green/Red wire that went into the red connector that plugged into the safety module, pull it out of that connector and change the end to a male bullet, and plug it into the extra Green wire that is hanging near the rear of the battery box. Presto, it becomes a K0-K2! Just don't forget to pull in the clutch or find neutral before starting, or you'll make the DOT mad... 

[Grinnin]

To thicken the plot, I found the following post from May 2013:

It looks like my SMSU has a diode that has failed open.  It'll still work without 2 of the 3 diodes but it becomes more susceptible to damage from voltage spikes.

This photo shows 3 pins soldered just right of center with no component drawn on them.

One of the capacitors shown in the Honda Update could fit on those pins,  The other 2 capacitors don't fit there.

From the Honda Update '74 dealer info.
Now anyone can build their own!

[mystic_1]

This photo shows 3 pins soldered just right of center with no component drawn on them.

Yeah I had noticed those extra solder pads also. 

I also note that the 550 schematic above doesn't state a year, and the photos of the disassembled units don't seem to show any capacitors.

Perhaps that 550 schematic is from a different year?  The various parts lists show that part number 31800-341-701 was used on the CB750K3-K5 and the CB550 (years not stated on the ones I checked, so I''d assume 1973-1975).  Perhaps there are different versions out there, or the later versions included the caps?

mystic_1

[CycleRanger]

Zombie thread rises!
Ok, so I've been kind of obsessing over these units lately, mostly because I needed a good one.
(Yes, I know no one really needs one.)

Anyway, since I have managed to collect four of these that don't work I thought I would open one up like Lucky did and see if I could get some better images of the components with the ultimate goal of building one these from scratch. (Why? Why not!)

The first thing I noticed is when I look at this I see three diodes, one resistor, one capacitor, and three transistors, assuming the small, three-pole component with the Y on top is a transistor and assuming all three of the black of little black things are diodes, I suppose one could be a resistor.
Which doesn't line up with either known schematic....

This unit came off a '74 550 and except for the heat sink it appears to match the one Lucky cracked open.
Unfortunately none of the components have readable markings (except the resistor) so I''ll have to guess at values which shouldn't be too difficult, right?
I tried getting readings on the components but results were mostly inconclusive.
I have three more of these I could crack open to compare but it's a huge pita to remove the epoxy without destroying the whole thing.
Anyway, if anyone has any thoughts, wisdom, or snarky commentary to contribute please step up.

[mrmunsell]

My two cents as an EE whose regular work does NOT involve circuit design (so take this with a grain of salt):

It looks to me like…

D1 -> prevents Q1 turning on when the clutch switch is closed, may also prevent the neutral light from getting (dimly) illuminated through this box when the clutch switch is closed.
D2 -> protects the box against reverse polarity (battery connected backwards, etc).
D3 -> protects this box from the relay coil back EMF.
R1/R2/Q1/Q2 -> allows a small current through the neutral switch to provide a higher current ground for the starter relay circuit.  The difference in currents depends on the transistor specifications (beta, etc)… but could be orders of magnitude.

So at least one purpose the box serves is reducing the current through the neutral switch (vs just using a diode).  Is that THE only purpose... or is it accomplishing anything else?  I couldn't say for sure... but no other purpose has occurred to me yet.  Judging by the size of that resistor and power transistors in CycleRanger’s images, I would guess the starter relay coil draws a fair amount of current.  Maybe that plus the neutral light current is too much for the neutral switch design.  That is just speculation since I don't know how much current the starter relay coil draws nor how much the neutral switch can handle… but I doubt Honda would add this module unless it was really necessary.

In CycleRanger’s images, I think that small transistor is Q1 and the two large power transistors are connected in parallel to act as one higher power Q2.  The traces / components that I could discern looked consistent with the diagrams in mystic_1’s post (in particular, I was looking at the one in Reply #12).

You probably already know this… but if you want to measure the values of any of those two lead components you need to desolder, cut, or otherwise disconnect one lead first.  Otherwise, for example, if you measure the resistance of a resistor while still connected, you are measuring the resistance of that resistor in parallel with the rest of the circuit.

[CycleRanger]

In CycleRanger’s images, I think that small transistor is Q1 and the two large power transistors are connected in parallel to act as one higher power Q2.  The traces / components that I could discern looked consistent with the diagrams in mystic_1’s post (in particular, I was looking at the one in Reply #12).

Good feedback. It hadn't occurred to me that the two power transistors are in parallel. That makes more sense.
(Although I just now noticed that the one in the photo Lucky posted apparently only has a single transistor.)
I am planning to desolder everything but this thing was already wrecked before I opened it up.
I'm not an EE but I spent 28 years in the semiconductor industry and did a lot of board troubleshooting and repair.

[mrmunsell]

It hadn't occurred to me that the two power transistors are in parallel. That makes more sense.
(Although I just now noticed that the one in the photo Lucky posted apparently only has a single transistor.)

Yes, I see what you mean.  That one also has a heat sink.

It could be that the revision of the box in that photo is using a power transistor with a higher power rating (when used with a heat sink).  And/or if anything outside the box changed that reduced the amount of current box needs to handle (such as a starter relay / coil revision), then the transistor current and power requirements would also be reduced.  Etc.

Also, I could not make out any capacitors in your images or Lucky’s.  In both, I believe I see two resistors though.  The high power one I referenced in my prior post (mounted on end) and a lower power one (mounted flat).

I am planning to desolder everything but this thing was already wrecked before I opened it up.
I'm not an EE but I spent 28 years in the semiconductor industry and did a lot of board troubleshooting and repair.

Cool… so it sounds like that task is right up your alley then.

Been awhile since I read a resistor color code, but I think that flat mounted resistor is (or perhaps was) 470 Ohms. 
Harder to make out the bands in Lucky's image... but it looks like at least three of the bands are consistent with yours (1, 2, and 4).

[CycleRanger]

There was a heat sink between the two power transistors on my unit.
Yes, I agree, Lucky's unit is likely a later revision that uses a single, higher power transistor OR they decided the two-transistor design was overkill for the application.)
(Actually, I can see the holes in Lucky's PCB for the missing second transistor.)

I don't think anything changed as far as the specs for external components between '74-'75, but any slight changes probably wouldn't have been publicly documented anyway.
This is how I interpret the components.

[mrmunsell]

I'm pretty sure the upright green thing is a capacitor, although, yes, it could be a big resistor.
This is how I interpret the components.

One reason I believe that is a resistor is based on your third image in Reply #19.  You can see the Ohm symbol... that was probably the last character in the resistance marking.  Another is that from what I could discern from the images, its connection to the other components appeared consistent with one of the resistors in that schematic.

This is how I interpret the components.

I agree with all your other labeling in that image.