• Don't overlook our Welcome Package, it contains many links to important and helpful information about functions at VHT like posting pictures and sending PMs (private messages), as well as finding the parts you need.

    AD

Brand New CDI for CB400T/CM400T/CM400E/CM400C/CB450SC

Maraakate

Veteran Member
Joined
Jun 7, 2022
Total Posts
2,418
Total likes
566
Location
Lancaster, PA, USA
Thanks to a kind donation from Jim I received 5 broken CDI units as was able to use these to successfully reverse engineer the CB400/CM400 manual series CDI.

CURRENT STATUS 09-23-2023:

Now available for order. 20 cases this time to keep up with demand!
Added product variation when ordering CDIs. It will now ask you if have the 1978 year. This is to let me know if you require the bullet connectors or 2-pin canon plug for the blue and white wires.

Winter time project sale. All CDIs are $100 off (sale price $200) until March 31. Get that bike running before spring!

Special 20% discount for VHT Forum Members. Use code 'VHTMembers' at checkout!


  • Total price is $300 shipped to the US. International customers please E-Mail me to discuss options. Current lead time is 2 weeks as each one is built individually by me and tested dynamically with an actual stator/rotor.
  • DSS now has it available. You can get it here with your weekly DSS fix via this direct link or searching manually for the part number: https://www.davidsilverspares.com/CB450SC-NIGHTHAWK-1985-USA/part_392654/. I've informed DSS that it fits other models, so it should show up for CM400T/E/C, CM450E, and CB450SC. If it doesn't please let me know so they can fix it.
    It should be showing up for other model years, if it's not let me know so they can fix it.
  • Kits are discontinued due to user error. If you can convince me that you're competent and understand there is zero warranty/support then send me a PM. I will only offer this for VHT members.

You can order the complete plug and play unit here. Other ordering options include my ebay store https://www.ebay.com/usr/eggmatorch, David Silver Spares, direct PM on VHT, or via E-Mail.

I have had a few questions elsewhere as to why the cost is what it is. The metal case, which bolts up the same as the original is high quality and manufactured locally. It's even thicker gauge than the original and slightly taller to allow more room for the components, reducing stress on them.

I take no compromises on the quality of this unit. Each CDI is hand assembled by me, with high quality automotive rated components and Japanese Nichicon capacitors, just like the original. The harness uses high quality components from Matt over at Vintage Connections/Sparck Moto. This is not cheap junk, it will last as long as, if not longer, than the original unit. I personally use it on my own bike and wouldn't settle for anything less.

20220909_204212.jpg

-----------------------------

Original thread:
I've started to work on one that had the wires already cut close to the grommet. This time I tried a different strategy of using a 1000w heat gun and slowly breaking apart the potting compound. Once the top was exposed, went around the corners with a brass chisel to expose the PCB ends. Then heated the back of the unit for about 2 minutes and was able to lift it out. From here I have been able to clean up the back of the PCB for the traces.

Some interesting notes compared to the CM400A CDI (at least in this revision): The wires are all arranged in a "bus" like format so they're all right in one spot. On the 400A CDI they go wherever so wire lengths vary. There are 3 through holes that are not connected to anything. Unsure of the schematic, but may have been leftovers from the 400A safety lockouts or this board is also used in other models from the same era with additional safety features. The board is larger than the 400A CDI PCB but is much more simplified. There are 4 film capacitors on this board, compared to 2 on the 400A CDI. This PCB also had a plastic piece for protecting shorting out the PCB on the metal case. The 400A CDIs I've taken apart it appears they poured a few mm's of potting compound, waited for it to flash off, then laid the PCB down and finished pouring the rest.

Ignore the bubble on the large main cap, that's from the heat gun.

297687363_1549710025449004_8817214969630041336_n.jpg

296917340_386730453397329_1224392981882699280_n.jpg

297584648_1390631488090424_253101566959613456_n.jpg

296989104_432872932229190_1931674294604324628_n.jpg

20220809_112712.jpg

20220809_112720.jpg

Going out to get virgin paint thinner from autobody supplier to help remove the rest. Though, with this board I'm going to try this different approach and it may not be necessary. I see there is silkscreening with japanese characters on the board. I have friends who can read Japanese and can probably translate it. I'd like to see what it says.
 
Last edited:
That one is the "C" version, the 1st generation are marked "A", 2nd generation supposedly is "B" but I've never seen one. The only A's I've seen are on early 1978 CB400T1 & 2, '79 CB and CM's have all been the C version.
I sent you 4 C's and 1 A so it'll be interesting to see what the changes are from A to C
 
That one is the "C" version, the 1st generation are marked "A", 2nd generation supposedly is "B" but I've never seen one. The only A's I've seen are on early 1978 CB400T1 & 2, '79 CB and CM's have all been the C version.
I sent you 4 C's and 1 A so it'll be interesting to see what the changes are from A to C

You asked, and I figured why not.

Carefully removed the potting compound with heat gun and patience on the "A" version. So, there are no "bus" lines, just like the Hondamatic version. This means the PCB layout is slightly different. Board size is the same as the "C" model. There are these little coiled stand offs near a few resistors. Unsure of their purpose, yet. I was amazed to see silkscreen layers with component layout. Once I realized this I stopped using a thin screw driver so I wouldn't scratch the silkscreen markings off and switched to a plastic panel tool and a lot of heat.

Here's some pictures of the process.

20220809_125024.jpg

20220809_125925.jpg

20220809_130844.jpg

20220809_133142.jpg

20220809_133523.jpg

20220809_133700.jpg

20220809_134531.jpg

NB: I desoldered the large capacitor in order to remove more of the compound. So far every CDI I have opened the large capacitor has no markings. It's not a huge deal, as it's been established many years ago on the yahoo groups that a 1.0, 2.2 or 4.0uF will work just fine. It's probably a 2.2uF. And they're all so close in value it makes no real difference. Yet again the 5P4M and 2P4M markings are impossible to read. I always assumed the thinner washed it away, but just seems like the paint they used for the markings on them comes off very easily. Just like the old vacuum tubes and their markings.
 
Last edited by a moderator:
Two hours in the pot, pretty clean. I don't think it's worth leaving it in there longer as it will eat the resistor markings and cause me grief. Have to take care of other things, so I will start desoldering the components one by one and laying it out sometime tonight.

One very interesting thing. The mostly clean board is the "unknown model". The other one that is practically the same is from a CM400C. There is one difference. R4 is not used in the CM400C box, but it is in this unknown one. The resistors are different values. It may have been intended that the single resistor from the unknown one is the same value as R4 + R5 in the other one and they added this spot for substitution parts. Until I take a closer look at the resistance values I can only speculate at this point.

There is apparently only one thermistor on this board, labelled R1. Hondamatics get two.

I bench tested two of the caps with my old Heathkit Capacitor Tester IT-28 for fun. One was around 2.7uF, the other was reading around 3.3uF. So a 2.2uF value is fine. 400V is the correct value, as I tried pumping 450V to it and heard some pops. Good to know most of the components are practically the same.

20220809_155450.jpg
 
You're making really good progress. In the end this will offer another option for all the SOHC 400 owners and it's great that you're doing it here.
 
The layout of the A and C boards are quite different.

Yes. The main difference seems to be that they moved the harness to be closer together in a bus-like configuration in the C revision to make it easier to solder and not have a bunch of wires laying on top of the components and possibly to reduce wire stress. That would explain the layout being different. The component values for the capacitors and diodes are the same. I'm guessing the resistors probably are too, or are very close to the same values. For completeness sake I'll document what's different between them, but I'm pretty certain there's no real difference in terms of how it operates on the actual bike.

When I get everything documented, components ordered and some proto PCBs from China I'll send one for you to test, Jim. Once it's verified good you can send it back to me and I'll re-pot it in one of the original cases for you.

You're making really good progress. In the end this will offer another option for all the SOHC 400 owners and it's great that you're doing it here.

Thanks! This one will be more documented than the 400A. The "other forum" removed my thread which actually had a lot of pictures and where I was generally at with things. But, now you'll be able to see how I did it. I'll put this one up on Github as well for people who are curious on the schematic. I'm sure kbongos will have some great fun with it.

Another thing I noticed is that I'm pretty sure ND did not make the manual CDIs, but they definitely did for the Hondamatic. There is no ND markings in the silkscreen or the copper layer. On the 400A CDIs they put their logo in the actual copper layer on the backside of the board. It's kind of neat to see. I'll get a picture of it later. But, I wonder if TEC or someone else did this one. They also use crimped stand-offs for the wire harness on the manual CDI, but on the Hondamatics they directly soldered it to the board. The thermistor on the manual appears to be "dipped"/"rubberized" but on the automatics some had a small fiberglass cover over it (like the wire harnesses for the alternator), some had nothing. I just used heatshrink tubing. It's all just intended to prevent shorting out on other components which should never happen since it's potted anyways. Just interesting very minor differences.
 
Spent some time this evening to lift each component individually, write the colour codes down, then verify with multimeter. Interestingly, you will see that there are a few R values that just flat out do not exist on the silk screen or the board. They are labelled R1-R15, but there's really only 12. C2 does not exist on this board, there is silkscreen layer for it, and the pads have just been soldered shut. I looked on the A revision it's a 0.47uF 400V. Maybe they found out later it wasn't really needed and omitted it for cost savings, who knows?

TR1 is labelled 351 and measured out the same at room temperature.

At this point now I have to lay out the same board in KICAD and it should be able to produce a schematic from the PCB. I'll try and goof with this tonight. In the mean time, at this point just about anyone can use the pictures and my notes and do the same if they wanted to.

20220809_211341.jpg

20220809_215827.jpg

20220809_221331.jpg
 
Last edited by a moderator:
Spent some time last night laying out the board in KICAD. Partially complete, will likely need to readjust the component spacing a bit once the schematic is done. Should be done today or tomorrow then I can send the boards off to China. Please excuse the hacky layout of the schematic. Just joining components one at a time as I hold the original PCB in my hand and keep looking at the monitor :lol: will make it more readable when finished.

Capture.PNG

partial.PNG
 
Last edited by a moderator:
There's nothing hacky about what you're doing.

Thanks, but I know one of my EE buddies would have a heart attack :lol:.

Here is what I believe is the original circuit, need to quadruple check this later and send it off. Layout needs tweaked on the PCB. Also need to measure it, didn't do that yet.

Capture.jpg

partial.jpg
 
Need to do a few more checks tonight so I stop doubting myself, but good thing I checked as I forgot to connect the R7, D6, C4 network to Pin 2 of SCR2. Now fixed. Also made the board layout a bit more optimal. There's 4 mounting holes that aren't necessary, but there on the original and it's trivial to add it to the drill file in the PCB so I went ahead and added 4 mounting holes in case someone wants a kit.

I measured the board, the original is 81.1mm/3.22in L x 46.7mm/1.84in W. We can do a bit better than that, especially since there's dead space for components that aren't there. New dimensions are 2.62in L x 1.79in W.

Capture.PNG

partial.jpg
 
Nice work. Eagerly watching. I like smaller components. Especially when I try to cram all those plugs and wires in my headlight housing…
 
Looking at the board it appears C1 only has a connection to P4, no input.

Yes, the CDI design is for all the manual transmission SOHC CB/CM 400/450 models including the CB450SC
 
Looking at the board it appears C1 only has a connection to P4, no input.

Yes, the CDI design is for all the manual transmission SOHC CB/CM 400/450 models including the CB450SC

Hi Jim,

No, it runs to Pin 2 of the SCR like it should. It's hard to tell because the "Courtyard" footprint is not the same. This is the large cap. I've attached a picture with the two through holes highlighted in red with arrows so you can see.

Capture.PNG
 
Hi Jim,

No, it runs to Pin 2 of the SCR like it should. It's hard to tell because the "Courtyard" footprint is not the same. This is the large cap. I've attached a picture with the two through holes highlighted in red with arrows so you can see.

View attachment 16318
I was sure I was missing something there, now it makes sense.
 
STATUS UPDATE 08-10-2022:

* Reviewed the board again, I'm satisfied. Ordered 10 boards from China. Should be here in about a week. Paid a bit extra to get it sooner via DHL.
* Updated the Github repository with the Gerber files in case anyone is curious.
* Added Digikey CSV file for import to Digikey cart with components on the Github repository. Film cap sizes may be wrong on the board layout, but we'll find out if it needs to be adjusted.
 
For a little bit of fun, put the large cap on the old heathkit capacitor tester (which needs alignment but good enough for this purpose) it's reading about 4.4uF so the 2.2uF should be just fine. The 400V/0.047uF reads about what it should be.

20220810_224743.jpg

20220810_225424.jpg
 
Last edited by a moderator:
STATUS UPDATE 08-12-2022:

* PCB Manufacturer was confused that the top/front copper layer had pads but no traces. After many back and forth emails they get it now so production will start tomorrow or Monday. Should receive by Friday. I specifically make the boards this way because they do not charge extra for two layer boards. Having copper pads for the components on the top and bottom, as well as the traces on the bottom allows you to solder the components from top or bottom of the board. It also helps to stabilize the components in there. More surface area for solder to really hold those components in place. I want this for long term reliability since it will be on something that is exposed to the elements and vibration.
* Made the schematic more readable for people who aren't me. Updates are on the github.
* Lifted a leg of each component on the Revision A board so we can document it. I'll make a PCB layout of it as well. I don't intended to manufacture those as there's very little purpose to do so, but will be good to eliminate any ambiguity between revisions.
* I have not verified this, but I am certain the Japanese characters on the PCBs are the colour codes for the wires. On the 'C' revision since they're all right next to each other it's harder to tell. But, on the 'A' revision the characters are always next to a wire.

Capture.jpg
 
That one is the "C" version, the 1st generation are marked "A", 2nd generation supposedly is "B" but I've never seen one. The only A's I've seen are on early 1978 CB400T1 & 2, '79 CB and CM's have all been the C version.
I sent you 4 C's and 1 A so it'll be interesting to see what the changes are from A to C

Figured out the minor differences. Uploaded Revision 'A' component list to the github, but for the thread here it is:

C2 is a 0.047uF 400V film cap on Rev. A. It exists on silk screen on Rev. C. but is not on the board.

R2 is 1M on Rev. A. It exists on silk screen on Rev. C but is not on the board.
R3 is 22 on Rev. A. It exists on silk screen on Rev. C but is not on the board.
R4 is 3.3K on Rev. A. On Rev. C. it is 1.8K.
R5 is 56 on Rev. A. On Rev. C. it is 2.8K.
R15 does not exist on silk screen on Rev. A. On Rev. C. it is 1.2K.

R6 exists on silk screen for Rev. A and Rev. C but it is not on either board.

Diodes, SCRs, and thermistor are the same.

I will try and work on the schematic and board layout for Rev. A sometime soon so we can see if there's any real difference; but considering they all work the same, probably not. If there's any true resistance differences when added together (not smart enough to do this in context of the entire circuit, sorry) it was probably to compensate for something slowly killing the Rev. A boards or to deal with stator coils hurting the CDI. Could be any of those things, or even just simple cost cutting moves. I.e. found out it was unnecessary and removed some of them and replaced them with other resistors.

I did find out those tiny coils soldered on R4 and R5 just seem to help stabilize R4 and R5 on the board. They ohm out at 0. It was literally just coil slug and an additional "leg" to help lengthen the resistors slightly. I don't see any real benefit or understand why this would be done. Other ideas could be for heat dissipation (but they are already 1W resistors so doubtful), or RFI, but I doubt that too. I can't answer for sure as I'm not an EE, just a hacker.
 
Here you go, Jim. Documented Revision A as a PCB and Schematic for future generations. One thing I noticed besides what is mentioned previously is that there are 8 diodes, not 9. They added a diode for black/white tracer wire and removed resistor and cap around it in Rev. C.

As always, available on Github repo as well.

Rev_A_PCB.PNG
Rev_A_Sch.jpg
 
STATUS UPDATE 08-15-2022:

* PCBs have already shipped! DHL says Friday.
* Ordered the parts from DigiKey, also should be here Friday.
* Re-purposed one of the harnesses from these CDIs by trimming back the wire and putting new crimp connectors on it. Will use this for the test CDI for Jim.
 
This is a fascinating thread. Apologies for my ignorance, I'm assuming that you are building a copy of the original CDI unit, using modern components.

So, what actually fails in the original broken CDI units ie stops them working ?
 
This is a fascinating thread. Apologies for my ignorance, I'm assuming that you are building a copy of the original CDI unit, using modern components.

So, what actually fails in the original broken CDI units ie stops them working ?

Yes, building a 1:1 clone of the original circuit but using modern components such as AEC-Q200 rated components for increased reliability.

What usually fails is the large green film capacitor. After 30-40 years the dielectric breaks down and they become "leaky". This means that as more voltage is applied to them the capacitance increases higher and higher and doesn't filter properly and causes strange things like late timing advance (or none at all). Sometimes they just get leaky enough that they completely open up, leak their dielectric fluid onto the board and it will refuse to spark.

Actually, I opened up one of the CDIs yesterday to scavenge that harness out of it and this particular unit had the dielectric leak out on the back side of the PCB and had that distinct "fish oil" smell to it. The dielectric is corrosive so the inside metal of the case was rusted from this. That's more of an extreme case. According to Jim's notes this one had a late advance.

So far, every resistor I've pulled out of circuit to test and note it's value has been spot on. They haven't seemed to drift, at least at low voltage, but it's quite possible the resistors also drift in tolerance badly when the high voltage is applied across it.
 
Thanks, it's all black magic to me, I can't even begin to imagine how that circuit works, like where the data for the ignition advance is stored, how that relates to engine rpm etc. These days it'd all be in some sort of integrated circuit, all on a chip.

So, to make a spark on the CM400 engine, there's this circuit (CDI unit), a crank mounted sensor, and ignition coil(s) ?
 
Thanks, it's all black magic to me, I can't even begin to imagine how that circuit works, like where the data for the ignition advance is stored, how that relates to engine rpm etc. These days it'd all be in some sort of integrated circuit, all on a chip.

So, to make a spark on the CM400 engine, there's this circuit (CDI unit), a crank mounted sensor, and ignition coil(s) ?

I'm not an EE myself. But the circuit theory is explained in the FSM. Basically (and someone may correct me here as I could be wrong), the SCR works for triggering the advance and this is triggered by a certain voltage threshold that is generated from the coil which would increase in voltage as the RPM increases.

Attached is some pictures from the FSM on theory of operation.

CDI_Theory_Page1.PNG
CDI_Theory_Page2.PNG
CDI_Theory_Page3.PNG
 
Thanks for that. Amazed that Honda put all that detail into the shop manual. I had to Google what an SCR was, I can now almost kid myself that I understand what is going on here.

This would be an example of analogue rather than digital electronics, the way the spark advancer pulser works ?
 
Thanks for that. Amazed that Honda put all that detail into the shop manual. I had to Google what an SCR was, I can now almost kid myself that I understand what is going on here.

This would be an example of analogue rather than digital electronics, the way the spark advancer pulser works ?

It's sort of a hybrid analog and digital setup since the SCR is a transistor to act as a digital "switch". The coil voltage being sent to the CDI and eventually to the SCR is analog.
 
There was a lot of change going on during this time period.
The 1981 CM400A is significantly different than 1981 CM400T in regard to this ignition CDI.
Thank you Maraakate for all the hard work reverse engineering these CDI's and
making this readily available. I'm still trying to make sense of it.
Honda did try to provide some technical information, but the circuits are difficult to make sense of.

So I got my hands on a FSM for my buddy's Y81 XL250S. It has a AC CDI, but the advance is done mechanically in a pulse rotor.
In same FSM there is Y82 Addendum where Advance changes to electronic in CDI. I include some related pics.

Some other history trivia is https://en.wikipedia.org/wiki/Honda_CX_series 1982 CX500T had first fuel injection as well as
initial transistorized ignition system(1981 GL500).

xl250s_cdi_17-0_pg213.PNGxl250s_Y81_cdi_19-7_pg230.PNGxl250s_Y81_cdi_19-8_pg231.PNGxl250s_Y82_cdi_24-86_pg346.PNGxl250s_Y82_cdi_24-87_pg347.PNG
 
So Maraakate, here is me trying to make sense of it. I got started learning a little KICAD by trying to add some labels to your schematic. I'm not sure if this is good bad or ugly but it is just some effort to try and help understand this stuff. I still can't look at the circuit and go, ok, that makes sense. There are basic elements that make sense, but there is definitely some confounding stuff there. What are they doing with two AC1, AC2? Why do they have 2 advance signals, what I call 'AdvPul' and 'F-43'? Very strange indeed.
CM400T_CDI_Schem_KB1.PNG
 
So what I am calling AC1, AC2 I think is a coil on the stator to give us 300v to charge the cap. I speculate that this is 2 taps of the same coil, where AC1 is middle tap, AC2 is outer tap. In a doc file you have in git you have someone's notes about it that seem to indicate this is the physical arrangement along with notes about it maybe being high power/low power for different RPM's. I can't say if this is true, it make's little sense to me.

The two trigger signals it get's from pulser, I call F-15(idle), and F-43(adv). I think those are two different coils embedded in this large pickup that is in place around the rotor and picks up on magnet sticking out of rotor. They don't really say, and the pictures provided might lead one to believe it is only one coil with taps, but I don't think that is the case. I suppose it's possible, like a leading trailing edge thing. But I know I can use these two signals with a switch to my discount CDI and it switches fire from F to Adv. I also did check with timing light lately and the OEM CDI did progressively advance from F to Adv as I throttled somewhere around 2-4K. Didn't seem that smooth, but my CDI is over 40 years old ;)

On top of that they have what I call 'AdvPul', and that is yet another signal that is a coil on the stator somehow involved with the advance. I want to call it a bias. Very strange stuff indeed.
 
If you're referring to the RobertF PDF from the CM400A repo that may not be exactly the same.

I really don't know the exact function of each component on the PCB as I'm not an EE by any means. What I'm doing I consider to be "grown up legos" as I tell the kids. I'm essentially taking it apart carefully and rebuilding it exactly without the instructions :). I have no idea if there's two coils in the 400T and one takes precedence over the other and triggers one SCR over the other. I believe this is the case, but I really don't know.

In other news, I took apart a harness and will document it in the notes section on the repo for the length of the harness so you can get the parts from Matt.

Coil PVC 6mm - 3 1/4"
Green Coil - 7 3/4"
Yellow Coil - 7 1/4"

Blue and White PVC 6mm - 2 1/2"
Blue - 27 3/8"
White 27 1/4"

Alternator PVC 8mm (Matt doesn't have this size but I use 6mm on the CM400A harness and it fits fine) - 8 1/4"
CDI PVC 9mm - 21"
Black w/ White hole location in CDI PVC - 3 1/4"
Black w/ White - 7 1/4"
Sky Blue - 32 3/4"
Pink - 32 3/4"
Green Harness 32 3/4"
Brown - 33 1/2"

Yellow/Green Coil Connector - Female Latched 6.3mm 2-pin Yellow is Pin 1
Blue/White Connector - Female Latched 6.3mm 2-pin Blue is Pin 1
Brown/Green/Sky Blue/Pink Connector - Male Latched 6.3mm 6-pin - Brown is Pin 1, Green is Pin 2, Pin 3 and 4 NC, Sky Blue Pin 5, Pink Pin 6.
Black/White - Female Bullet 3.5mm

Grommet positions (from left to right)
Row 1 - Brown, Black/White, White, Green Harness
Row 2 - Blue
Row 3 - Green Coil, Yellow Coil, Sky Blue, Pink
 
On top of that they have what I call 'AdvPul', and that is yet another signal that is a coil on the stator somehow involved with the advance. I want to call it a bias. Very strange stuff indeed.

So according to the circuit theory page I posted above from the FSM, it does appear the fixed and advanced coil do bias. If I am understanding correctly, as RPM increases, more voltage is produced, which triggers the SCR faster and they reach a point where advance is limited. I.e. "governor" in car talk. I suppose it could be possible to play around with values on the PCB to adjust the curve but that's out of my league and I don't care that much either. I'd rather have it just be reliable instead of trying to squeeze every last bit out of it; but that's just me.
 
I want to ask about the 86' CB450SC Nighthawk as far as it being different than the 82-85' CB450SC Nighthawk models?
I was told this by someone when I tried to locate a used stator for it in the late 90's;did the CDi box get more digital with that model/year ?
 
I want to ask about the 86' CB450SC Nighthawk as far as it being different than the 82-85' CB450SC Nighthawk models?
I was told this by someone when I tried to locate a used stator for it in the late 90's;did the CDi box get more digital with that model/year ?

No idea what is actually different inside of it. Never worked on, been near, or owned an 86 CB450SC. I checked the fiche and apparently that CDI is unique to that year(!). If you have a broken CDI you want me to crack open to see if it can be reverse engineered (i.e. uses discrete components not custom ICs that would require decapping/x-ray/EEPROM readers) I would be happy to do so.

With that said, Jim said earlier that this should apply to the CB450SC. I look back at 85 and 86 years and they still have the same can and same looking stator. So it's probably the same, but they superseded part number for some reason. It might be some slightly different resistor values, etc. in those years. Again, if someone has broken ones from a specific year they want me to look at it let me know. It looks like 87 is when they made the move to surface mount (SMD) soldering and I am not equipped to build those kinds of boards... but it seems like every chinese vendor out there is focused only on those kind.
 
I want to ask about the 86' CB450SC Nighthawk as far as it being different than the 82-85' CB450SC Nighthawk models?
I was told this by someone when I tried to locate a used stator for it in the late 90's;did the CDi box get more digital with that model/year ?
No, the stator and CDI are exactly the same as the earlier years/models. Honda did a part numbering change over in 1986 using letters instead of numbers for the middle 3, probably running out of number combinations.
 
No, the stator and CDI are exactly the same as the earlier years/models. Honda did a part numbering change over in 1986 using letters instead of numbers for the middle 3, probably running out of number combinations.


Hi Jim,
I tried getting a used stator assm. back in the late 90's when I had the 86' CB450SC Nighthawk and salvage yards told me it was different.
They are different part numbers.

What you're saying is,the CDi box is the same ?
 
Hi Jim,
I tried getting a used stator assm. back in the late 90's when I had the 86' CB450SC Nighthawk and salvage yards told me it was different.
They are different part numbers.

What you're saying is,the CDi box is the same ?
Yes, that's what I'm saying. The CDI is exactly the same as all the other SOHC CB/CM 400/450 manual trans models. Here's the list from CMSNL https://www.cmsnl.com/honda-cb450sc...lifornia_model7221/unit-assy-cdi_30400mc9831/
CB400T HAWK 1980 (A) USA
CB400T HAWK 1981 (B) USA
CB450DX 1988-1993 BRASIL
CB450P 1988-1993 BRASIL
CB450SC NIGHTHAWK 1982 (C) USA
CB450SC NIGHTHAWK 1985 (F) USA CALIFORNIA
CB450SC NIGHTHAWK 1986 (G) USA CALIFORNIA
CB450SC NIGHTHAWK 450 1983 (D) USA
CB450T HAWK 1982 (C) USA
CM400C CUSTOM 1980 (A) 1981 (B) USA
CM400C CUSTOM 1980 (A) 1981 (B) USA
CM400E 1980 (A) USA
CM400E 1981 (B) USA
CM400T 1979 (Z) USA
CM400T 1980 (A) USA
CM400T 1981 (B) USA
CM450C 1982 (C) CUSTOM USA
CM450E 1982 (C) USA
CM450E 1983 (D) USA

This product fits to 19 models.

The only reason you don't see the 1978/79 CB400T1 & 2 is the Blue/White connector on those is a bullet instead of a spade plug.

 
Crazy, yet not crazy, that the CB450 was still being made in Brazil until 1993. I say not crazy as I used to collect a lot of Sega Master System stuff years back and they were still making it in Brazil way past it's life expectancy. If I would have known about that it's probably possible to go on Mercado Libre (Brazil Ebay) and find a CDI from 93.

It's a moot point now that I've reversed engineered it.
 
STATUS UPDATE 08-17-2022:

* Boards arrived. Look good. Soldered a harness on one of them for Jim while I wait for parts.
* Digikey says parts should arrive tomorrow. Will solder it up and ship it Friday to Jim for testing.

20220817_124431.jpg

20220817_130206.jpg
 
Last edited by a moderator:
Back
Top Bottom