Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

transistor to mosfet issue

Status
Not open for further replies.

jrbe

Newbie level 5
Joined
Dec 14, 2010
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,366
I am trying to repair an old standalone ECU. I had 2 bad ignition transistors out of 4. I google searched the # on the transistor but came up with nothing. I dont have the old ones in front of me but can add their part # in later.

A 0-3.3v signal was going in but nothing was coming out of the 2 dead transistors. all 4 transistors had the same signal in but only 2 were working. Their grounds and pull up voltage and traces all checked out ok. I contacted the company to see what they replace the transistors with because the old ones were discontinued. They recommended IRFIZ48V (pdf datasheet link in 3rd post below, sorry, posting rules...) which are way overkill but its "what they use because we have them in stock" I bought some and installed them by their directions but now the new mosfets are all not fully pulling to ground on all of the channels now.

A simplified diagram of how i have the mosfet wired in the ecu and important stuff to the waveform is (in my 3rd post below)

There is a 0-3.3v signal that makes a nice square wave that turns the mosfet on while the ecu wants to fire the coils (image in 3rd post) which is the same on all 4 channels

the ignition output (out of the mosfet) holds at ~5v until the ecu sees all the crank and cam signals then starts firing the coils. Thats what the long pull to ground is, but then only pulls from the 8v to about 7v which isnt enough to fire the coils. (image in 3rd post) Its strange it will pull to ground before the ecu wants to fire the coils but then only pull to 7v when it wants to. The 8v is steady and constant the whole time. It runs a hall sender that is used as a sync (cam/reset) signal that is working

Anyone have an idea whats wrong?

---------- Post added at 08:42 ---------- Previous post was at 08:41 ----------

2nd post so i can post oscilloscope traces to help the above wording

---------- Post added at 08:52 ---------- Previous post was at 08:42 ----------

Or not. if someone could reply to this i can post links to the scope waveforms to help see whats going on.
 

have you tried to find a similar transistor out of cook books your self
think you should find a suitable replacement your self
the transistors similar to IRFIZ48V are: 2SK1257,2SK1653,2SK1952
is your original transistor one of the above ?
if not i think the problem might be with the replacement
 

Yes, i tried crossing them with no luck. The transistor? that came out has K940 as a part # and 5D on a second line on 1 and 41 on the other 3, guessing the second line is a batch or date #. I wasnt able to find these anywhere, a datasheet for them or anything that had them listed to cross to. They are supposed to be very high speed transistors.
IMG_20101214_104024.jpg


irfiz48v datasheet pdf

This is how the Mosfet is wired in a simplified ecu diagram
dualsportigncircuit.jpg


0-3.3v signal (blue trace volt markings on the right)
dualsport0-3v.jpg


8v driver out signal, the pause is waiting to see the crank and cam sensors to get the coil sequencing and timing right, thats the pull to ground which is strange to me because it wont do it once it wants to fire the coils.
dualsportnotpullingtoground.jpg


Sorry for the sorta large pics.

---------- Post added at 11:57 ---------- Previous post was at 11:06 ----------

the 3.3v comes from the microcontroller through a 470ohm resistor to drive the transistor normally if that helps
 

Thanks for the info. How do you know that there should be 2S at the beginning of the part#?

Im an electronics noob, sorry. I cant read chinese or whatever that datasheet was but i found 2SK Semiconductor (Transistor, diode, IC) Cross reference but looking at this the pins seem backwards unless im reading the diagram backwards. From the pic i posted of the k940, the bottom pin gets the 0-3.3v signal, center is the output (0-8v) and the top is ground. Does it make sense that these are 2sk940's with this info?

I also tried to find the 2 cross reference parts 2SK975, 2SK1274 at digi-key, mouser, and newark. Cant find them. Are there any other sources to find uncommon electronic parts?
 

i have this old cook book that has lots of stuff in it
the similar parts for your transistor are :2SK975 & 2SK941
 
  • Like
Reactions: jrbe

    jrbe

    Points: 2
    Helpful Answer Positive Rating
I found a place that stocks the 2SK975 and ordered a bunch. Thanks everyone for their help. Hopefully these take care of it.
 

Ok, cool, thanks for the info.
 

Just wanted to give an update. Those replacement 2SK975's dropped right in and worked like they are supposed to. Its making a nice 0-8v square wave again now.

I do have a new problem though. The way this ecu triggers the coil is by grounding the outputs that have a ~200ohm pull up resistor to a regulated 8v. The coils i am trying to fire need a ~5+v signal when on and float when off.

So I need essentially the same spec mosfet as the 2SK975 but in a p-channel and i have to rewire them a bit. How do i go about finding the same specs but just the opposite channel type mosfet? Im probably not wording this exactly right but i think you know what i mean.

I tried to search digikey and newark but even with the spec sheet in front of me digikey and newark find nothing close.

The switching speeds are very important to keep timing right. Anyone know a good mosfet finder site or a way to find the opposite channel of a mosfet you have?

Thanks in advance.

Jay
 

Hi,
I don't have the data book with me right now, so won't be able to tell the exact part. I'll see if I can find it later on. Anyway, if you have a data book with MOSFETs listed, look for MOSFETs starting with 2SJ. These are generally P-channel MOSFETs suitable for use along with 2SK series MOSFETs.

Hope this helps.
Tahmid.
 
Awesome, that helps a lot. At least i have a direction to look now.

Thanks again for the info!
 

try this one
2sj 148
thanks to Tahmid for the the advise
 
  • Like
Reactions: jrbe

    jrbe

    Points: 2
    Helpful Answer Positive Rating
Im pretty sure the 2sj148's are powerful enough for what i need to do. The one thing im not sure on is the gate on voltage. I was under the impression p-channel mosfets needed a 0 to negative __ voltage to turn on, opposite of a a 0 to +3.5v of how its running the n channel mosfet.
All i see is + power on these, i dont get how they can turn on.

So i have a ground(0v) to +3.3-3.5v signal for the gate turn on voltage. This is gate threshold voltage correct? Or is it 5v (supply) minus the signal (gate) to turn on the p-mosfet? The 2SJ148's have a -2 to -3.5 Vth (gate to threshold voltage.) Will these work?

So besides changing the source to 5v and getting the pull up resistors off the drain is there anything i need to do to run these off the microcontroller? Between the microcontroller and the mosfet, there is a 470ohm resistor.

Thanks to everyone for the help, i appreciate it!

Jay
 
Last edited:

Hi,
Connect a pull-up resistor of about 1k from gate to source.
The gate-source voltage mentioned in the datasheet(as well as the threshold voltage) is the voltage at the gate compared to the source (negative[Vg-Vs]). Eg. If the threshold voltage is 3v and your supply is 5v(applied to the source), applying anything less than (5-3) = 2v(with reference to ground) [putting it the proper way with reference to S, 3-5 = -2v(with reference to source)] will turn the MOSFET on. There is also a limit to the maximum Vgs you can apply.

Hope this helps.
Tahmid.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top