30 yr old ECU capacitors measurements? Replacement type brand voltage and temp ?

[Rango]

Hi guys. So I have a question regarding ECU, car's computer. My car been jerking at certain amount of rpms and shows code 9 which leads to distributor. Dizzy been replaced twice. Nothing fixed.

Long story short. I decided to open up ECU and check capacitors. Car is from 1992 so it 30 yrs old at this point. Caps visually look good and no burns anywhere i can see on the board.
Board looks clean to my eyes. I started measuring capacitance of caps. The car has total of 9 aluminum electorlytic capacitors, round brown cylider shapped. All 105C temp rating KME BRAND.

Default value of MM capacitance reading with cables not conntect to anything 0.300nf aka control reading. ECU was not connected to any power supply.

C14 220uf 35v measures 0.269nf=269uf default value of MM. My MM shows ~0.269mf when it can not measure it. This appears to main capacitor that power the board? It's biggest cap on the board!
c15 33uf 35v measures by flashiing on/off value of 10.22uf. Open circuit?
c24 47uf 10v measures 60.3uf ,way over capacitance
c21 33uf 35v measures 33.1uf appears to be good
c62 10uf 35v measures 0.269nf=269uf default value of MM. Can not get any other reading.

c19 33uf 35v measures 0.399nf=399uf default value of MM. Can not get any other reading.
c18 220uf 10v measures 0.397nf=397uf (another nf reading instead of uf). default value of MM. Can not get any other reading.
c95 100uf 10v measures 101.32uf appears to be good (will replace anyway due to age)
c71 10uf 35v measures 0.312nf default value of MM or value of cables not connected to anything. Can not get any other reading.

So am i reading this data correctly that only 2 capacitors out of 9 are working? Even board's main is down?
How is car even running with 7 caps that are down or at best malfuctioning? C14 appears to be main cap powering the board as it's biggest value cap!?

Can you provide any insight what is going on with those caps based on what i've posted? What type (alum or ceramic?) and brand of caps should i replace them with?
Any perticular brand with lowest % tolerence levels?
I've also seen one of those caps i can only find in 85C rating, not 105C. Should i looks at other vendors that carry 105C. Also are ceramic caps, orange disk shapped better for this car application?

There are also yellow squared capacitors or diodes with labels of 104J & 124J pictured next to brown onces.
Should i measure and replace those as well? Not sure what to compare values to? Also any idea where i can find schematic for this ECU? Besides those labels no other values are printed on them.

Any advise will be much appriciated! Thank you kindly.





yellow squared capacitors or diodes with labels of 104J & 124J pictured next to brown onces to the right of the pic

[james_s]

The yellow squares are film capacitors, those are good quality and very rarely fail, leave them alone.

Replace the electrolytics with good quality 105C or higher rated parts, I like Panasonic, Nichicon and Rubycon personally.

[bdunham7]

What make/model?  Acura Integra perhaps?  Or some Acura/Honda model?

The circuits may be getting in the way of proper testing.  You would need to either desolder the capacitors and retest or use a special low-voltage tester that can work in circuit.  It looks pretty easy to desolder these, so that's probably the way to go.  I wouldn't be surprised to see one or more of these bad.

There's no need to worry about tolerances for replacements, but I would try and find AEC-Q200 rated units.  Mouser/Digi-Key/etc should list that characteristic.

[wraper]

There is almost no chance you will get normal capacitance reading within the circuit, especially when measuring with multimeter.

[Rango]

It's 1992 Honda Civic VX aka EG6

So you're saying resistor may be getting in a way of charging that cap so therefore MM shows bogus or no reading?

That makes sense as i doubt car would run with 7 out 9 caps begin down.

[bdunham7]

That makes sense as i doubt car would run with 7 out 9 caps begin down.

They could be pretty badly off and the ECU would still mostly sort of work.  Those have all linear regulators and such and the capacitors don't work all that hard normally. 

Are those new/reman/tested/junkyard distributors or what?  The crank sensor issue is/was pretty common, you probably need to do some serious diagnostics on the circuit all the way from the CKP inside the distributor through all the wiring and then back into the ECU.  You'll need an oscilloscope for that. 

[Rango]

That makes sense as i doubt car would run with 7 out 9 caps begin down.

They could be pretty badly off and the ECU would still mostly sort of work.  Those have all linear regulators and such and the capacitors don't work all that hard normally. 

Are those new/reman/tested/junkyard distributors or what?  The crank sensor issue is/was pretty common, you probably need to do some serious diagnostics on the circuit all the way from the CKP inside the distributor through all the wiring and then back into the ECU.  You'll need an oscilloscope for that.

2 dizzys were bought from store, maybe were remanufactured. VX has specific one too which i made sure it was for VX specifically D15z1 engine.

Crank sensor sounds familiar to my ears as far as me reading about on forums. Initially my car died 5 yrs ago and mechanic replaced dizzy. He was really stuggling with figuring out what was wrong.
He got it running but i think he patched and never really resolved the issue, cause that jerking never went away.

Car runs ok now with exception of having weird jerking at specific rpms. It doesn't sall or shut down or anything like that. I'm thinking maybe one of the caps in ECU is bad is is not acting right but i could be way off too. I'm just guessing.

I've also noticed that 1 or 2 spark plugs are coverd in oil so i will need to replace value cover gasket. The oil there maybe causes misfires and therefore throwing code 9 which also is recognized as engine misfire in cylider, which would makes sense as spark plug is covered in oil. I will also be doing value adjustments at the same time but wanted to see ecu caps condition.

I should replace all those 9 caps anyway right? I was also thinking that maybe one of two of those caps is not working therefore causing weird issues with car and throwing that code 9.

The problem with that code 9 is when i look it up it's described that could be caused by wide veriaty of causes and since mechanic had problem figure it out i thought maybe it's 30 yrs old capacitors in ECU.

[james_s]

Oh, it didn't even occur to me that you were trying to measure capacitance in-circuit, that will never give you a reliable measurement. Capacitance rarely tells you what you need to know anyway, you need a meter that can measure ESR, that's what typically rises when capacitors fail. ESR can oftem be measured in-circuit,

[Jeff eelcr]

Those caps go bad with time and temp.
Most all I have seen have phyisically leaked by now so I
never measured them first. Panasonic NHG is a good replacement.
Also note that is not the only modual with cap problems.
Jeff 

[Rango]

Those caps go bad with time and temp.
Also note that is not the only modual with cap problems.

Event tho they physically may look good, inside they may not be, meaning values could be off already?
Physically they look fresh, no leakage anywhere on the board that i can see.

Btw i should go with those radial shape not disk shaped orange ceramics right?
Those ceramic onces i'm reading they don't leak but maybe not be good for automotive application.

What other modules in car have caps jeff?

[Ice-Tea]

Those caps go bad with time and temp.

Event tho they physically may look good, inside they may not be, meaning values could be off already?
Physically they look fresh, no leakage anywhere on the board that i can see.

Btw i should go with those radial shape not disk shaped orange ceramics right?
Those ceramic onces i'm reading they don't leak but maybe not be good for automotive application.

Looks don't count for much and there goes your proof.

You probably won't find ceramics for those high bulk capacitance values but even if you did you shouldn't use them. Sometimes, a circuit depends on a certain amount of ESR to be stable, to reduce inrush or whatever else. Replace like with like.

[JPortici]

I don't think i've ever had to replace caps in an ecu, or issues related to caps anyway (not even the big bulky caps used on the injector boost converter in diesels). diodes/power switches, on the other hand...

[free_electron]

KME is not a brand !. These are nichicon caps KME Type.


https://www.westfloridacomponents.com/mm5/graphics/S06/KME250VB10RM10X20LL.pdf

[Rango]

KME is not a brand !. These are nichicon caps KME Type.


https://www.westfloridacomponents.com/mm5/graphics/S06/KME250VB10RM10X20LL.pdf

Oh, thank you. I've added to my cart nichicon Low impedance onces, instead of general purpose as they have double hours of use vs gen purpose. That's ok right?
There is also higher temp onces, 135C but not for all the value selection and they're double the prices. I don't live in arizona so maybe overkill?

[Kleinstein]

Measuring the capacitors in circuit is a bit tricky. It can work in some cases, but it may also show a wrong value (especially too large) from the rest of the circuit. So much too small would be suspicious and may be a replacement.
The electrolytic caps are mainly at the supplies and there quite high tolerance (e.g. +200%/-70%) are still acceptable and the circuit may even work with some of the capacitors missing or open circuit all togehter. I somewhat doubt a capacitor is causing the error message.

The blue drop like capacitors may be worth a check too: they look like tantalum caps and these may fail short. The yellow film caps don't age much.

[mariush]

KME is not a brand !. These are nichicon caps KME Type.


https://www.westfloridacomponents.com/mm5/graphics/S06/KME250VB10RM10X20LL.pdf

Oh, thank you. I've added to my cart nichicon Low impedance onces, instead of general purpose as they have double hours of use vs gen purpose. That's ok right?

It's united chemi-con / nippon chemi con KME series, and it's a decent one.

If you decide to replace, replace with any other series that has the specs about the same : look at maximum esr and maximum current ripple columns for the capacitance and voltage rating of capacitor .. esr could be slightly lower but don't go with ultra low esr and I'm not sure I'd replace them with solid/polymer capacitors... just don't know if they rely on the capacitor ESR for something and replacing with a capacitor that has barely any ESR could be bad. 
Maximum current ripple can be higher.
If the diameter and height allows, usually you can go with higher voltage rating capacitors that may be in a bigger package (usually bigger packages means better specs)... for example instead of 35v rated you could go with 50v rated capacitors.

There is also higher temp onces, 135C but not for all the value selection and they're double the prices. I don't live in arizona so maybe overkill?

Rubycon has an OK series, RX30, rated for 130c ... for example here's
10uF 50v rated at $0.49 each, in big enough quantity (11k) : https://www.digikey.com/en/products/detail/rubycon/50RX3010M8X11-5/3563311   ... the 10uF 35v kme  has 41 mA current ripple, this series 10uF 50v can handle 200mA

Plenty of 33uF  50v and 63v rated as well, same diameter, $0.6 each : https://www.digikey.com/en/products/detail/rubycon/50RX3033M8X11-5/3563302

[Gyro]

It's curious that manufacturers of cheap consumer gear feel the need to smother their capacitors in white 'snot' (that goes brown and conductive with age), yet there isn't a hint of it in that ECU! It must be a reflection of PCB quality rather than lead fatigue.

Most of the electrolytics look to be stood off the board and have wider footprint lead spacing than the capacitors. You should probably try to mirror the mounting heights reasonably closely and form the leads properly. At least use a pair of fine nose pliers between the capacitor seals and the bends to avoid stressing them. Don't just spread the legs.

[Jeff eelcr]

 I have been replacing caps in ECU and other moduals from 1987-1989 to about 1996-1998
when most started to be manufactured with Panasonic capacitors depends on vehicle.
Florida has a tempature that causes them to fail sooner than other areas of the world.
 I have had several corroded so bad that replacement was required.
Jeff

[Jeff eelcr]

 Rango if any modual in any vehicle is received the first thing we look at is capacitors
well 2nd thing first is if its potted if potted we just return them... most of our customers are
car dealers, mostly engine or transmission but have seen many others, mostly japanise cars.
Caps I see failing any with con in the name as in ruby, nichi, united chem... so I do not use them.
1989 to 1995 is most of the problem years and yes only the electrolytic caps.
Jeff

[mariush]

Do you realize you're complaining about 25-30 year old components which have electrolyte in them which degrades with use over time, which is sealed inside the capacitor with a rubber seal that after 20+ years could be all cracked?
It's a component with limited life... It's like complaining that incandescent bulbs break eventually.

Also, about that time ('99 - '04) there was a problem with capacitors being made with a bad electrolyte formula (rumors are an engineer stole incomplete or incorrect electrolyte formula from Rubycon) ... see  https://en.wikipedia.org/wiki/Capacitor_plague
 

[Jeff eelcr]

This problem started showing up about 1990-1991 in Bose OEM amplifiers
using Nichicon capacitor mostly, manufactured starting about 1987.
The only capacitor company I talked with that knew of and did something
was Matsushita (Panasonic) The KA, KG series were designed not to have
that problem. The NHG series was already a replacement for an earlier
series that outgased to much in both cases the seal material interacted
with the electrolyte so I was told then.
And yes capacitors are rated for time also, hours per degs C not years.
I have seen these Bose units unused sitting on the shelf leak so these
caps are going to go bad 100% sometime.
Funny I worked on a 67 volvo radio today bad power switch, caps OK.
Jeff

[Rasz]

There is a mayor difference between hot power amplifier and Honda ECU.
Those caps look fine, doesnt mean they are fine, but they do live in an optimal environment - low power low temperature. You can replace them, but Im giving it less than 1% chance of fixing your issue.

>Dizzy been replaced twice. Nothing fixed.

has it been diagnosed? or just replaced because idiot light came on and instruction pointed at the distributor?
is it misfiring? Does it smoke(blue) when started cold? Have you checked ignition timing using the light? checked plugs/leads?
https://honda-tech.com/forums/honda-civic-del-sol-1992-2000-1/92-civic-vx-d15z1-stumbles-40-50mph-then-throws-code-9-a-3244289/

[Jeff eelcr]

The amp I am talking about is in vehicles in the same enviroment
that the ECU is in and does not run hot.
The heat comes from the sun here inside vehicle can be very hot.
Summer months are bad, here its 99% chance that those caps are bad.
5-10 years ago we saw them weekly maybe 10 at a time sometimes and
replacing those caps and repairing open runs caused by caps leaking,
fixed the problems. (from car dealers and electrical repair shops)
However we only repair vehicle electronics not vehicles.
Jeff

[Rango]

Rubycon has an OK series, RX30, rated for 130c ... for example here's
10uF 50v rated at $0.49 each, in big enough quantity (11k) : https://www.digikey.com/en/products/detail/rubycon/50RX3010M8X11-5/3563311   ... the 10uF 35v kme  has 41 mA current ripple, this series 10uF 50v can handle 200mA

Plenty of 33uF  50v and 63v rated as well, same diameter, $0.6 each : https://www.digikey.com/en/products/detail/rubycon/50RX3033M8X11-5/3563302

Mariush big thanks for going extra mile and pointing those rx30 series. I think i'm going to go with those since they're nearly the same prices as the Nichi i've added to cart.
I'm thinking that i will be replacing all those caps anyway since ECU is what reports codes so if its brain is scrambled by bad caps then no amount of mechanical repair will fix the issue.
Especially that those caps are literally pennies and any other mechanical component is pricey and mechanics are especially pricey.

[james_s]

Well it's probably not a terrible idea to replace the capacitors but I will say I'll be surprised if it is the problem. Most issues like this I've seen are caused by things like dirty connectors or bad sensors.