• 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

CD250U CDI

ashimotoK0

Veteran Member
Joined
Jun 7, 2020
Total Posts
107
Total likes
5
Location
East Yorkshire UK
My friend in the UK has lots of big Hondas but himself and his partner both have their cherished CD250U's, which they ride regularly.

One of their CDi's packed up and after various aftermarket units and used genuine units failing of not even working, I am tasked with de-potting the original, plotting out the circuit , diagnosing the fault and then repairing/refurbishing.

Luckily, the potting is only semi-hard (probably polyurethane) ...in the past, as part of my 'day job' I have had to de-pot hard epoxy encapsulants using a boil up in a cocktail of solvents including T.H.F. (tetrahydrofuran). This is quite aggressive and can destroy plastic coated parts and their markings, so I decided to not use solvents and just a blunt screwdriver and a kebab wooden skewer.

After several hours whilst watching TV and picking away, I am down to the main PCB, identified and measured most of the component values and gathering data before I sit down and trace out the circuit. BUT before I do this wonder if anyone on here has already traced it out? The unit uses an integrated circuit (LD/MB4213) which is still available from China (see attached for datasheet) for the pulse conditioning, thyristor triggering, advance timing etc. The main thyristor (SCR) which dumps the high voltage capacitor charge into the ignition coils is an NEC 3P4MH , which is still available too. There is an additional small thyristor which I suspect may be to do with some kind of limiter or cutout but all will be revealed once I have a circuit diagram. I have not found any component failures yet but there were a few soldered joints which may have been 'dry'. I plan to replace the parts anyway with uprated ones (capacitors particularly), I may even do a new epoxy-glass PCB, as I have been designing and making these most of my career.

Last time I did something like this I was hurled abuse on the CX500 forum with the usual unhelpful comments ... 'why bother ...anorak/nerd, MacGyver etc' twaddle. I have been an engineer in the electronics industry for 40 years and I am always intrigued by these things and the IC they used in this CDi is a new one on me, so I am doing it because I want to ...so no negative comments please.
 

Attachments

  • cdi chip.JPG
    cdi chip.JPG
    41.6 KB · Views: 22
  • cdi depot.JPG
    cdi depot.JPG
    86.5 KB · Views: 19
Last time I did something like this I was hurled abuse on the CX500 forum with the usual unhelpful comments ... 'why bother ...anorak/nerd, MacGyver etc' twaddle. I have been an engineer in the electronics industry for 40 years and I am always intrigued by these things and the IC they used in this CDi is a new one on me, so I am doing it because I want to ...so no negative comments please.

Well, you know this isn't the CX500 forum and we don't bash people who try to repair or replicate factory parts like CDIs that are scarce, in fact we absolutely encourage this kind of thing. Have you seen Frank's (VHT member Maraakate) work on the SOHC 400 series CDI replacements? We encouraged him to come here to develop and promote his effort because that other forum, where he started originally, took down his thread after he refused to give them $100 a month for advertising.

https://www.vintagehondatwins.com/f...C-CDI-Reverse-Engineering&highlight=maraakate
 
Like Tom I have to point you towards Frank since he's done both the manual and A model CDI's for the SOHC 400/450 series

With respect, I read though that and that CDi is a totally different design to the one used in the CD250U sold here in the UK ...it is more like a Superdream one from the early 80's, the CD250U is late '80's. I am particularly interested in the ASIC chip used on the CD250U so I will carry on. Pretty sure my friend has tried a CDi from another earlier 400 model and it didn't work. There are three high voltage capacitors on the CD design and dual (two separate) coils even thought it's a 360 degree crank, Type 2 engine.
 
With respect, I read though that and that CDi is a totally different design to the one used in the CD250U sold here in the UK ...it is more like a Superdream one from the early 80's, the CD250U is late '80's. I am particularly interested in the ASIC chip used on the CD250U so I will carry on. Pretty sure my friend has tried a CDi from another earlier 400 model and it didn't work. There are three high voltage capacitors on the CD design and dual (two separate) coils even thought it's a 360 degree crank, Type 2 engine.

Interesting, and I'm sure Frank would be happy to offer some suggestions as he knows a pretty fair amount about the reverse-engineering process.
 
Hi!

I saw 250 and CDI in the topic title and couldn't resist :lol:. Yes, I reversed engineered the 400 units.

In your case, it looks like you have figured most of it out already. So kudos to you! It looks like the MB4213 is now considered obsolete by vendors. You can check around, there may be ones with similar specs that will work the same. Infineon may have a recommendation on a drop-in replacement. Sometimes they just supersede the part numbers with a ROHS compliant design, you might get lucky.

If you have to get NOS of that IC make certain it does not come from China on eBay. If you get it from eBay, also avoid California. Usually, Chinese vendors will lie and say it's in Cali or drop ship to Cali. There's a huge problem with Chinese vendors sanding off the markings of other ICs then silkscreening whatever on it. Just make sure you're getting it from a reputable vendor that's not from China.

Does yours have a thermistor? Both 400 designs had at least one. I assume to help with changes in temperature for cold starting. I've just used somewhere around ~3400-3500K B value thermistors. I never actually calculated the B values on those. Considering the circuit, I don't think it's mission critical to have the exact same temperature curve. The bikes are not that sophisticated. As long as it's close it should be OK.

Diodes are probably 1N4002, but if you're worried try 1N4004s. It will work all the same. In the 400 design they used both, probably as a cost cutting move, but you could use 1N4004s in the whole thing and would work.

You say nothing looks wrong on yours... in my experience it's the yellow film caps and/or solder joints that cause the units to go bad. Usually the large yellow cap (which maybe yours does not have one or you desoldered for pics?), but I see no reason why the smaller ones can't go bad. It looks like that blue one is a tantalum, yes? If so, it was thought they last practically forever like ceramics but we're finding out now with a lot of 80s equipment that is not true. Resistors almost always are in spec when I've measured them out.

If this uses the same type of compound that was in the 400Ts you should be able to remove most of it very carefully with a heatgun and a plastic body panel tool. Work slowly. Cleaning up the last remnants on the back side you can use Mineral Spirits/Wax and Grease Remove or Paint Thinner. Be very careful if you use Paint Thinner if there are silkscreen markings you are trying to document as the thinner may wipe it right off. Take lots of pictures.

Are you willing to take pictures of the back side? I'd be willing to work with you to produce a brand new module. I'd even give you a prototype for free.
 
Did very minor research for you... Digikey claims this to be a drop-in replacement: https://www.digikey.com/en/products/detail/analog-devices-inc-maxim-integrated/MAX314EPE/1512678

Specs are slightly different, but I think it will probably work all the same. If you get the board working again you should try and change the IC to have a socket and give it a try. Let me know if it works for you.

Try and replace the film caps and just reflow solder. That's likely all it is. It's possible the SCR/Thyristors got hosed if one of the caps shorted. They're fairly cheap to replace so might as well just do them too. I wouldn't bother with the resistors unless it's obvious one is bad.
 
Wow .. what fantastic replies to my post. I have been away in the sun for a week but next week I plan to trace out the circuit and photo the PCB foul layout. I have not seen any thermistors yet but I know the CX500B one I de-potted had them and IIRC I abstained some with the same curve. I got a lot of negative press on the CX500 forum and it kind of refocused me ...I found a CX with only 300 miles left in a damp shed for 35 years and the moisture had got down to the PCB ..there were voids in the potting (Hitachi tuttut) exposing bare PCB joints. I sold the bike a guy totally restored it to standard...not sure if he retained the rebuilt CDi ..probably not. I will share all of my findings on here ...I have no wish to reproduce myself so anyone who want to take up the baton is welcome! I am just helping a good friend out.
 
Sounds good to me. I'm looking to expand into other Honda models that are unobtanium that aren't SMD (I'm not equipped and have no experience in it either).

What do you use for PCB layout? I use KiCAD. I post my work on Github with Digikey parts lists that you can feed into the cart as a CSV: https://github.com/maraakate

Yes, the Hitachi CDIs seemed to use some sort of soft compound. The ND ones from the 400A were some rock solid thing that crystallized. That took a few weeks of babying in virgin paint thinner to be able to start working on it.
 
Back in UK after a week in Tenerife. To be honest I don't design many PCB's these days, as I have 'universal' PIC and analog designs from years back that usually I can adapt. But I use a UK based company package called EasyPC ...it's basic but produces all of the necessary Gerber files for manufacture (silk screen, solder mask, multi layer, drill NC etc.) Once I have plotted out the circuit of the CD250U CDI, I am interested to see what the purpose of the second (small) thyristor (SCR) is for .. I suspect it's maybe some kind of rev. limiter or cutout interlock ..but that's purely supposition at the moment. The potting appears to be the same as the CX500 Hitachi one...i.e. a semi hard polyurethane I would think. When I used to reverse engineer competitors products for my work, 'potted' with the hard black epoxy potting I used to boil up in tetrahydrofuran and MEK mixture but this was very aggressive and dissolved the cap. coverings etc. I once remember getting a truck module from a show in Texas and when back in the UK I de-potted the hard black encapsulant ..I also fished out about half a pound of gravel used to bulk up the expensive potting, I guess. This was when I worked for a UK company Fenner who had a fluid power/hydraulics factory in Rockford Il. .. which became SPX.

Don't be afraid of SMD .. I use it all of the time (for more than 25 yrs) .. there are simple ways of hand soldering even the tiniest of multi-legged chips using a 'rolling ball' of solder and lots of re-work flux. I did see that the chip in the CD250U module is (was) available in a 14 pin SOT package.

Such a breath of fresh air on this forum to allow tech posts like this ...with none of the nastiness of many others I could mention. Thanks guys ! :)
 
Back in UK after a week in Tenerife. To be honest I don't design many PCB's these days, as I have 'universal' PIC and analog designs from years back that usually I can adapt. But I use a UK based company package called EasyPC ...it's basic but produces all of the necessary Gerber files for manufacture (silk screen, solder mask, multi layer, drill NC etc.) Once I have plotted out the circuit of the CD250U CDI, I am interested to see what the purpose of the second (small) thyristor (SCR) is for .. I suspect it's maybe some kind of rev. limiter or cutout interlock ..but that's purely supposition at the moment. The potting appears to be the same as the CX500 Hitachi one...i.e. a semi hard polyurethane I would think. When I used to reverse engineer competitors products for my work, 'potted' with the hard black epoxy potting I used to boil up in tetrahydrofuran and MEK mixture but this was very aggressive and dissolved the cap. coverings etc. I once remember getting a truck module from a show in Texas and when back in the UK I de-potted the hard black encapsulant ..I also fished out about half a pound of gravel used to bulk up the expensive potting, I guess. This was when I worked for a UK company Fenner who had a fluid power/hydraulics factory in Rockford Il. .. which became SPX.

Don't be afraid of SMD .. I use it all of the time (for more than 25 yrs) .. there are simple ways of hand soldering even the tiniest of multi-legged chips using a 'rolling ball' of solder and lots of re-work flux. I did see that the chip in the CD250U module is (was) available in a 14 pin SOT package.

Such a breath of fresh air on this forum to allow tech posts like this ...with none of the nastiness of many others I could mention. Thanks guys ! :)

I can do some hand soldering of SMD. If you want to do any kind of modifications to circuits (especially old video game consoles) then you have to learn that part of it.

That's pretty cool you worked at a place that was reverse engineering stuff so you've got experience in this. I've never boiled it with actual chemicals, just water. That works OK on softer stuff. I'm assuming Tetrahydrofuran and MEK isn't available to a mortal like me in small quantities, needs proper ventilation, etc. so best forget about it?

Interesting about someone using actual gravel as a filler!
 
Such a breath of fresh air on this forum to allow tech posts like this ...with none of the nastiness of many others I could mention. Thanks guys ! :)

Ash, we would be stupid to discourage anything that a member can do to help the community with one of the more problematic and costly aspects of our favorite bikes. Hey, some of us are opinionated (myself included) but we're not dumb enough to "cut off our own nose to spite our face", as grandma used to say. :)
 
Did very minor research for you... Digikey claims this to be a drop-in replacement: https://www.digikey.com/en/products/detail/analog-devices-inc-maxim-integrated/MAX314EPE/1512678

Hi Frank (I am Ash BTW) .. that chip comes up as a quad switch not an ignition ASIC ? Did I get the link wrong please?

I started plotting out the circuit this afternoon, around the ASIC chip and it's looking like a similar circuit to the data-sheet schematic, albeit with some component values different. So far-so-good that I can measure and identify the markings on the passive components but I need to go to work Monday to measure the capacitance values with my bench TTi DMM. Quite often with de-potting you can damage a resistor etc, rendering it unidentifiable.

Once I have a full circuit I will draw a proper schematic on OrCad Capture (very old version but OK for circuit diagrams) and post it here as a .PDF file.
 
Hi Frank (I am Ash BTW) .. that chip comes up as a quad switch not an ignition ASIC ? Did I get the link wrong please?

I started plotting out the circuit this afternoon, around the ASIC chip and it's looking like a similar circuit to the data-sheet schematic, albeit with some component values different. So far-so-good that I can measure and identify the markings on the passive components but I need to go to work Monday to measure the capacitance values with my bench TTi DMM. Quite often with de-potting you can damage a resistor etc, rendering it unidentifiable.

Once I have a full circuit I will draw a proper schematic on OrCad Capture (very old version but OK for circuit diagrams) and post it here as a .PDF file.

Ahh, disregard. If you type in LD4213 into digikey it pulls up ALD4213 which is not the same thing.

eBay has 100pcs left of MB4213 apparently. It's from Singapore. It's possible AliExpress or someone else might legitimately have more.

At that point, with a custom IC, might be worth having someone decap it. I have no experience in that sort of thing, maybe you do?
 
Sorry for multiple posts, but if you're quite confident on the design I'd be willing to gamble buying the ICs and sending you some or trying to go through AliExpress/AliBaba/etc. to locate the IC for you.

I'm also curious. Is there a harness out of the other end of the PCB or is it some kind of special pin connector? If it's a pin connector, might make things more difficult to reproduce unless they just took some off-the-shelf connector (think Atari and Sega and Vectrex using DB-9).
 
Sorry for multiple posts, but if you're quite confident on the design I'd be willing to gamble buying the ICs and sending you some or trying to go through AliExpress/AliBaba/etc. to locate the IC for you.

I'm also curious. Is there a harness out of the other end of the PCB or is it some kind of special pin connector? If it's a pin connector, might make things more difficult to reproduce unless they just took some off-the-shelf connector (think Atari and Sega and Vectrex using DB-9).


con-2.JPG con-1.JPG
 
Last edited by a moderator:
Well, I am now down to a bare PCB and half way through doing a full circuit diagram, using OrCad Capture. I will put this on here in .pdf form, once it's done.

Poor quality PCB material (not epoxy glass) and all leads bent over including the IC, so not the easiest of boards to desolder. So far, I have not found any faulty components, so I guess the fault(s) were down to dry joints. If I was reproducing I would use FR4 epoxy-glass PCB material and plated thru holes to avoid dry joints. I will probably order up new parts apart from the two thyristors (SCR's) and rebuild the board, just to get my friend up and running again in the short term.

CDi_pcb -f and foil.JPG
 
Yes, the CM400A ND branded boards had the same issue. I use FR4 for the CDIs I sell and plated through holes for the same reasons you stated.

The way I was able to get the board down enough to reverse engineer with same type of crappy compound was to use virgin paint thinner. I'd let that sit in a covered jar for a day, check on it's progress, peel more off and keep going. Once I got done far enough then I started using a heat gun and a plastic body panel tool to chisel away the rest. Amazingly, that was slow enough that it didn't destroy any components and I was able to repair that particular one.
 
Can you purchase boards with 'plated through holes' ? Or does it require the builder to create that on their own ?
 
Just a quick update (particularly for Frank)

I have now plotted out the schematic (circuit diagram) for the Cdi using OrCad capture software and ordered all of the capacitors and resistors to refurbish the original PCB. I will put a .pdf copy of the schematic on here, once I have double checked it.

Most of the resistors are 1/8 watt,wire ended and tiny metal-film, which I had great difficulty sourcing in small quantities here in the UK, so I opted for surface mount 0.4W 'MELF' capacitors, which are cylindrical, very small and have metal end caps. These will mount on the underside (copper foil) of the PCB. I didn't want to fit larger resistors, mounted upright (like Hitachi did on the CX500 Cdi) because this would be mechanically bad, unless you are using a plated through hole PCB, which the CD250U PCB isn't.

One stumbling block though, is that I think two of the original electrolytic capacitors on the original PCB, I am fairly sure have been the wrong way round, since manufacture. I really need a second opinion on this from Frank but obviously I need to present my findings on here first.
 
this is what I mean about two capacitors being the wrong way round (reversed polarity)

Picture_20230220030230.jpg Picture_20234321104333.jpg
 
Last edited by a moderator:
Hi Ash,

Been very busy past couple of weeks. Unknown why that cap polarity is backwards. When you rebuild, fix it and see what happens maybe? Otherwise, I can try and ask an EE friend that would know more about it than me as for the reason why it might be that way. What does it say under the cap on the silkscreen?
 
One final thing, on the back I see the positive side of that cap leads to nothing (except that ground bus). It's copper is missing. Was this on purpose or did it flake off during your depotting? The reason I ask is that I have worked on arcade machines where there are caps where one side is connected and the other is not. So this may be the case, but considering it looks like part of the copper trace ripped off I doubt it, but want to ask for clarity.
 
This is a fascinating thread, even though most of it is straight over my head.

Slightly off topic, I've always thought that the CD250u would be a nice classic to own, wire wheels and disc front brake, with a modern(ish) engine. Am I wrong in thinking that this is essentially the same engine as used in the CB250 Nighthawk, and if so, could the CDI from one of those could be used ? There seem to far more of those around than CD250u's.
 
Well this ended up in smoke and tears today.

I plotted out the circuit and rebuilt the unit with all new passive parts and had a new potting box 3D printed

I don't have the bike here to test, so I built up a rig with a 60v AC supply from a toroidal mains transformer, used a CB500F ignition coil and an inductive type proximity sensor to simulate input pulses. Connected up .. got tiny HT sparks out of the coil ..understandable as a very low AC voltage compared with alternator CDi generator coil and the wrong type of ignition coil.... Then bang ! the smaller of the two thyristors dramatically went up in smoke then taking two of the power diodes with it. I think the problem may be that the AC generator coil on the bike can only supply a piddly little current, as the kill switch and ignition switch (latter when off) short that generator coil to ground. I think the small thyrisor is some kind of crowbar circuit and couldn't handle the high amp current of my mains transformer ...silly me. Not sure my circuit diagram is correct in that region of the circuit.... I need to re-check it. This video clip shows the dramatics https://www.youtube.com/shorts/3H8VhDwVJAk

One positive though is that I removed all of the blown parts and fed the CDi chip from a 30v DC bench PSU via a 2k7 limiting resistor and connected an LED with limiting resistor to the output, which drives the main CDi thyristor. I fed the pulser input with a winding from a stepper motor to generate suitable pulses, when the stepper motor shaft is turned and this worked fine, proving that the original I.C. (now socketed), works fine. (see this video clip https://www.youtube.com/watch?v=OiP__HZnZXY )

I am still not sure of the function of the smaller thyristor which blew up ..perhaps it's some kind of limiter and perhaps can be omitted. It's the one part of the assembly that I couldn't identify it's part number !:cry:

cdi cct cd250u.jpg cdi in pcb assembled.jpg cdi in potting box.jpg
 
Last edited by a moderator:
Sure glad you guys know what you're doing (no sarcasm meant at all). I feel prehistoric reading all of this.
 
What thyristor did you try? Do you have a pic of the original? On the 400s they used 2P4M and 5P4Ms. I'm pretty sure the 2P4M was a cost cutting move and you could use the 5P4M with no issues.

2P4M modern replacement: https://www.digikey.com/en/products/detail/ween-semiconductors/BT151-500R-127/1154785
5P4M modern replacement: https://www.digikey.com/en/products/detail/stmicroelectronics/TS420-600T/1040214

Do you have a circuit or parts list of how you simulated the bench test? This is something I'd like to have to make my life easier with testing these before I send them out.
 
What thyristor did you try? Do you have a pic of the original? On the 400s they used 2P4M and 5P4Ms. I'm pretty sure the 2P4M was a cost cutting move and you could use the 5P4M with no issues.

2P4M modern replacement: https://www.digikey.com/en/products/detail/ween-semiconductors/BT151-500R-127/1154785
5P4M modern replacement: https://www.digikey.com/en/products/detail/stmicroelectronics/TS420-600T/1040214

Do you have a circuit or parts list of how you simulated the bench test? This is something I'd like to have to make my life easier with testing these before I send them out.

Thanks for reply Frank. The thyristor, which went pop was the small one (like in a TO92 case) it is used in some kind off protection circuit mode I think. I really need to re-plot that area of the circuit and check I got it correct. I think the main thing with testing is to have an AC high voltage supply (like 200 volts) which is current limited. I suspect that the generator winding for the HT that feeds the CDi with AC voltage can only supply a small current and can withstand it's winding being short circuited briefly, because the Kill switch on the bike basically shorts out that coil. The stepper motor using one of it's coils is good as a pulse generator. The main CDi thyrisor is like the one you listed and is the original one, as it tested out OK. I did get a manufacturer part reference of the main thyristor but the module is at work and i need to check it when i am back on Tuesday. I replaced all of the recifier diodes with BYW 56 's as we use them at work

https://www.mouser.co.uk/ProductDet...rVV_ytJWPmyOwlbINyHDC15un2Vdh6xBoC2z4QAvD_BwE

Richard (Pitman) ... sorry for late reply ... my friends have two CD250U's and love them.... the only thing they lack is a kickstart.
 
Frank .. and anyone else interested.. I checked CD250U Shop Manual and it gives a resistance around 300 Ohms for the 'CDI Exciter coil).

So I am going to build a test setup which is 110-200 VAC limited by a large power resistor around 300 Ohms to simulate the exciter coil.. It also needs a switch to simulate the kill switch and ignition switch, on the bike, which will effectively kill the AC voltage into the CDi when the switch is closed by shorting out the exciter power. I am wondering if an AC mains polyswitch fuse of say 200mA could be used to automatically limit the current in the resistor (when the kill switch is closed) so that the resistor only needs to be around 40 watts for the worst case of 200 VAC exciter output (or is it more than 200 VAC at full revs .. dunno).

Problem is I don't yet know the wattage of the CDI or the maximum /minimum voltage from the Honda exciter coil at tick-over and maximum revs.
This is atypical polyswitch AC fuse (PTC) https://docs.rs-online.com/bdbc/A700000006480682.pdf

My circuit is sketched out in the attachments

Exciter cd250 .jpg exciter sim cgi.jpg
 
I now think that the circuitry associated with the smaller thyristor (SCR2) is a voltage controlled crowbar circuit. The circuit detects the rectified DC derived from the exciter coil and if below a minimum voltage crowbars the voltage to ground.
 
Interesting. So it sounds like for the 250U things are quite different. The CM400s use a true AC CDI throughout the entire circuit.
 
Interesting. So it sounds like for the 250U things are quite different. The CM400s use a true AC CDI throughout the entire circuit.

Problem is the small thyristor went pop and I have no idea of the spec. I just bought a couple of 400v 800mA ones in the same TO92 package type MCR100-6 Plus I don't really know at what AC voltage the exciter coil goes active into the main circuit. i.e. at what voltage input the small thyristor stops conducting to ground.... Also I don't have the actual bike here to scope test/measure the exciter coil output. Bit of a nightmare really. It's basically working though because with the small thyristor out of circuit I am getting sparks at the test ignition coil when I simulate ignition pulses with a stepper motor coil and about 60 VAC connected to the modules exciter coil input. I suspect it really needs around 200v. Once I have a unit running I can scope test the bike (the owner has two of these bikes .. a 'his and hers' situation.

This is the smaller TO92 thyristor I ordered:- https://datasheet.octopart.com/MCR100-6G-ON-Semiconductor-datasheet-9729370.pdf
 
Sounds good, keep me posted. I was looking over your diagram for a test bench and excuse me for some dumb questions (I'm not an EE, but mostly a hacker) but the 40W power resistor at 300ohm is the exciter coil (i.e. the coil that generates the voltage?), yes? For the pulser are you actually removing a pulser coil from a flywheel or have you simulate this somehow? What's the specs on the stepper motor?

What I was planning on doing at some point is to get a hosed CM400T and CM400A motor and take the top end off and leave only the crank and flywheel and spin the flywheel with a high power drill to simulate. I think this would work just as well, yes?

Thanks
 
Sounds good, keep me posted. I was looking over your diagram for a test bench and excuse me for some dumb questions (I'm not an EE, but mostly a hacker) but the 40W power resistor at 300ohm is the exciter coil (i.e. the coil that generates the voltage?), yes? For the pulser are you actually removing a pulser coil from a flywheel or have you simulate this somehow? What's the specs on the stepper motor?

What I was planning on doing at some point is to get a hosed CM400T and CM400A motor and take the top end off and leave only the crank and flywheel and spin the flywheel with a high power drill to simulate. I think this would work just as well, yes?

Thanks

Hi Frank .. a bit of a delay at the mo' as my wife's car was rear ended and probably a write off but at least she is OK. So sorting that mess out is my priority.

The exciter coil is the high voltage coil on the alternator, which produces the high voltage AC for the CDi. I fitted the resistor because the Honda CD250U exciter coil has quite a high internal resistance and hence is inherently current limited. The small SCR (thyristor) is used to clamp this voltage (I think !) at low output and relies on the AC source being current limited (around 300 Ohms according to the Shop Manual) . If you use a mains transformer to simulate this exciter voltage and it can provide high current, then the thyristor burns out when it is in it's clamping mode. Hence the limiting resistor I added to prevent this.

I reckon one phase coil of any decent size permanent magnet type stepper motor can be used to simulate the ignition pickup coil pulses. I would probably have used a a rotating magnet if I had the Honda pulser part but I don't have one. I am currently using a function generator on sine wave setting to simulate the ignition pulses.

Your suggested setup using actual components would be a useful way to go if you are testing a lot of units but not feasible cost wise in my case.
 
Didn't think of using a function generator for the pulses. Good idea.

Sounds like the 250 CDI is a bit different. On the 400s there is a low and high RPM coil. On the 400As they crossover somewhere around 50-70V. I guess they did away with this on the 250s and it's just one coil? To simulate it on the bench I suppose you measure the coil resistance and then use the large power resistor to simulate the same on the transformer?

What about using a variac or would that burn the variac out?
 
Well ... I fitted the rebuilt CDI to my friends bike and fingers crossed, it seems to run OK (he took it for a 10 minute run up to around 70 mph) so he's potted it in re-enterable soft silicone rubber potting compound.

In the end I put a socket on the PCB for the small unidentified thyristor (SCR) that went pop due to my initially harsh test method.
I tried running without the SCR fitted and the engine started, ran and cutout when turned off/kill switch operated to off, fine. So I left it out of circuit. I reckon it's function is to act as an AC nder voltage clamp, meaning that the exciter coil has to reach a certain voltage before the AC output is presented to the input capacitor/rectifier arrangement. Problem is that the small SCR had no markings on it after I had removed the potting around it, so I don't know it's precise characteristics.

I plan to investigate this low-voltage clamp circuit further but I wanted to get my friend mobile again ASAP. i have bought three different TO-92 package devices to test.
 
Well ... I fitted the rebuilt CDI to my friends bike and fingers crossed, it seems to run OK (he took it for a 10 minute run up to around 70 mph) so he's potted it in re-enterable soft silicone rubber potting compound.

In the end I put a socket on the PCB for the small unidentified thyristor (SCR) that went pop due to my initially harsh test method.
I tried running without the SCR fitted and the engine started, ran and cutout when turned off/kill switch operated to off, fine. So I left it out of circuit. I reckon it's function is to act as an AC nder voltage clamp, meaning that the exciter coil has to reach a certain voltage before the AC output is presented to the input capacitor/rectifier arrangement. Problem is that the small SCR had no markings on it after I had removed the potting around it, so I don't know it's precise characteristics.

I plan to investigate this low-voltage clamp circuit further but I wanted to get my friend mobile again ASAP. i have bought three different TO-92 package devices to test.

Nice progress. What SCRs are you looking at?
 
Bit of not so good news as copied in italics below... got this back from my friend at the weekend :rolleyes: I had to re-use the input capacitor (2.2uF) from the old unit as I couldn't find a replacement physically small enough. Also, socketed the IC and wondering if silicone potting affected the junctions. Gonna order new chips, input cap. and make a prototype PCB in epoxy-glass material. Determined to crack this one !

"Just did a 100 mile test ride.
50 miles on your CDI and 50 miles on the CDI off my bike. I wanted to do a perfect comparison so took the same roads in a circuit.


Your CDI results :-

Sluggish acceleration.
Doesn't pull up hill, have to change down one and two gears.
Top speed on flat is only 55mph. Down hill top speed 59mph.

The CDI unit from my working bike, results. :-

Crisp acceleration.
Pulls up the same hills on the circuit without changing down gear.
Top speed on flat 65mph.
Down hill top speed 70mph.+"
 
Bit of not so good news as copied in italics below... got this back from my friend at the weekend :rolleyes: I had to re-use the input capacitor (2.2uF) from the old unit as I couldn't find a replacement physically small enough. Also, socketed the IC and wondering if silicone potting affected the junctions. Gonna order new chips, input cap. and make a prototype PCB in epoxy-glass material. Determined to crack this one !

"Just did a 100 mile test ride.
50 miles on your CDI and 50 miles on the CDI off my bike. I wanted to do a perfect comparison so took the same roads in a circuit.


Your CDI results :-

Sluggish acceleration.
Doesn't pull up hill, have to change down one and two gears.
Top speed on flat is only 55mph. Down hill top speed 59mph.

The CDI unit from my working bike, results. :-

Crisp acceleration.
Pulls up the same hills on the circuit without changing down gear.
Top speed on flat 65mph.
Down hill top speed 70mph.+"

The Nichicon 2.2uF should fit, but it's currently discontinued and intended to be replaced with a new number. Vendors couldn't get me lead time because I believe Nichicon has not told anyone when it's being released.

The closest I've found that would probably fit is the one Cornell makes PN: 225MWR400K.

You can also try a 4.0/4.4uF. They're quite large, but I was able to make them fit in my cases. I don't think the value of the capacitor is critical as long as you are within ~1-4uF and at least 400V rating. Out of all the ones I pulled, they were all 2.2/400 but one had a 4.0/400. The circuit was the same, so I think they just used whatever they had.

As a test, I built a prototype with a 4.4uF and drove around. Worked exactly the same. So I doubt the capacitor is the issue, but maybe it is with the IC?
 
"Just did a 100 mile test ride.
50 miles on your CDI and 50 miles on the CDI off my bike. I wanted to do a perfect comparison so took the same roads in a circuit.


Your CDI results :-

Sluggish acceleration.
Doesn't pull up hill, have to change down one and two gears.
Top speed on flat is only 55mph. Down hill top speed 59mph.

The CDI unit from my working bike, results. :-

Crisp acceleration.
Pulls up the same hills on the circuit without changing down gear.
Top speed on flat 65mph.
Down hill top speed 70mph.+"
What's being described sounds like when the CB/CM 400 looses it's advance. Classic CDI failure for those.
 
I believe the 400 circuit the additional SCR is for the advance/hi-rpm side. So it may be possible with that missing it's being restricted.
 
Back
Top Bottom