dead vnd5n07

Testing and development of Megasquirt 3

Moderators: jsmcortina, muythaibxr

Peter Florance
Super MS/Extra'er
Posts: 3653
Joined: Fri Apr 02, 2004 8:40 pm
Location: Virginia Beach, VA
Contact:

Re: dead vnd5n07

Post by Peter Florance »

SymTech Laboratories wrote:
Peter Florance wrote:Can you measure the gate to source resistance of the dead device?
We have a dead VND5N07 here, and the gate-to-source resistance is ~49k ohms, just like the working ones; so the status feedback circuit isn't enabled if that's what you're asking. It's interesting to note that the MOSFET's body diode seems to shorted: the Rd-to-s and Rs-to-d are both in the 200k ohm range. Perhaps there's a negative voltage at the drain causing the diode to conduct (e.g. flyback recovery overshoot).
That is what I was wondering. The body diode fault is interesting.

I've killed a few but they were all clear abuse on my part. :(
Peter Florance
PF Tuning
81 BMW Euro 528i ESP Car
60-2 Wheel LS2 Coils, Low Z Inj
Co-Driver 1999 BMW E46 DSP car.
SymTech Laboratories
Super MS/Extra'er
Posts: 2188
Joined: Sun Aug 31, 2008 4:02 pm
Location: South Florida, USA
Contact:

Re: dead vnd5n07

Post by SymTech Laboratories »

Peter Florance wrote:That is what I was wondering. The body diode fault is interesting.

I've killed a few but they were all clear abuse on my part. :(
This particular VND5N07 was driving a single high impedance (14 ohm) fuel injector when it failed, so it wasn't subjected to much abuse at all.
SymTech Laboratories, LLC ----- MSQ File Repository: megasquirt.symtechlabs.com
Image
Peter Florance
Super MS/Extra'er
Posts: 3653
Joined: Fri Apr 02, 2004 8:40 pm
Location: Virginia Beach, VA
Contact:

Re: dead vnd5n07

Post by Peter Florance »

SymTech Laboratories wrote:
Peter Florance wrote:That is what I was wondering. The body diode fault is interesting.

I've killed a few but they were all clear abuse on my part. :(
This particular VND5N07 was driving a single high impedance (14 ohm) fuel injector when it failed, so it wasn't subjected to much abuse at all.
Agreed.
Peter Florance
PF Tuning
81 BMW Euro 528i ESP Car
60-2 Wheel LS2 Coils, Low Z Inj
Co-Driver 1999 BMW E46 DSP car.
EFIDART512
Helpful MS/Extra'er
Posts: 46
Joined: Wed Mar 24, 2010 10:00 am

Re: dead vnd5n07

Post by EFIDART512 »

subwoofer wrote:
EFIDART512 wrote:I just scoped mine to look at the active clamp (internal to the device) to have a look. It clamps right at 39 volts.
If the internal clamp failed, then the device could/would get killed by the fly-back pulse from the injector coil. A secondary clamp in addition would provide protection if the active clamp died. Mine isn't running on the car yet but I imagine that a stuck ON injector isn't a good thing.

[snip]

Attached is a pic of the working vnd5n07 if anyone is interested.
Has anyone looked at it with a high resolution scope (like an Agilent 3000 or 7000)? There could be a short spike there that goes above 60V and over time wears out the protection circuit (or the FET itself). A slow sampling scope would miss the spike for sure. A transient protection diode may be faster than the internal circuit?
Not sure how fast the Agilent units are, my scope is 300Mhz.

The app note for the LM1949 peak/holder driver uses a 33V 5W zener diode for the fly-back and the 2N6044 has a Vce rating of only 80V. I would say that the FET in question could tolerate that much.
Has anyone seen the transistors fail in peak-hold circuits? Maybe Jean can comment.

IR makes a part the same as the ST VND5N07. I have tried it and it operates exactly the same. Maybe the IR part is better (LOL).
As mentioned, no one knows for sure what is killing the part.
Keithg
Super MS/Extra'er
Posts: 2413
Joined: Sun Mar 06, 2005 9:15 am
Location: Chicago, IL, USA
Contact:

Re: dead vnd5n07

Post by Keithg »

The Saab trionic uses MTD3055VL (same package). I can look at the board tonight and see what is connected to the drain other than the injector. These have been extremely reliable driving high impedance (16 ohm) injectors. I can also check the dead VND5N07 as well. It was pulling down the output of the fet driver, so it was showing ground on the gate and the injector was 'on', so it appeared to be shorted source to drain as well.

KeithG
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: dead vnd5n07

Post by racingmini_mtl »

Keithg wrote:The Saab trionic uses MTD3055VL (same package). I can look at the board tonight and see what is connected to the drain other than the injector. These have been extremely reliable driving high impedance (16 ohm) injectors.
That's interesting. I have used them with my 4-channel driver board in the same configuration as the VND5N07 and they were not reliable. If you look at the drain-source avalanche rating, you'll see that its significantly lower than the VND5N07.

I'd be interested to see what they use with it, if anything.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Peter Florance
Super MS/Extra'er
Posts: 3653
Joined: Fri Apr 02, 2004 8:40 pm
Location: Virginia Beach, VA
Contact:

Re: dead vnd5n07

Post by Peter Florance »

racingmini_mtl wrote:
Keithg wrote:The Saab trionic uses MTD3055VL (same package). I can look at the board tonight and see what is connected to the drain other than the injector. These have been extremely reliable driving high impedance (16 ohm) injectors.
That's interesting. I have used them with my 4-channel driver board in the same configuration as the VND5N07 and they were not reliable. If you look at the drain-source avalanche rating, you'll see that its significantly lower than the VND5N07.

I'd be interested to see what they use with it, if anything.

Jean
I'm guessing Saab is not depending on any internal clamping
Peter Florance
PF Tuning
81 BMW Euro 528i ESP Car
60-2 Wheel LS2 Coils, Low Z Inj
Co-Driver 1999 BMW E46 DSP car.
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: dead vnd5n07

Post by racingmini_mtl »

EFIDART512 wrote:The app note for the LM1949 peak/holder driver uses a 33V 5W zener diode for the fly-back and the 2N6044 has a Vce rating of only 80V. I would say that the FET in question could tolerate that much.
Has anyone seen the transistors fail in peak-hold circuits? Maybe Jean can comment.
I don't use the data sheet circuit. There is a 36V Zener but it doesn't carry the flyback current. And I use the TIP122 which is rated at 100V but they are Darlingtons not FETs. And the VND5N07 is more than just a FET so there are different things that could fail.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Keithg
Super MS/Extra'er
Posts: 2413
Joined: Sun Mar 06, 2005 9:15 am
Location: Chicago, IL, USA
Contact:

Re: dead vnd5n07

Post by Keithg »

The dead Omnifet reads 260 ohms d-s, s-d, d-g, g-d, g-s, s-g.

I cannot see anything other than the caps to ground on all connector pins. It appears that the drain is connected directly to the injector and there appears to be no zener or other component on the board associated with this trace.

Keith
EFIDART512
Helpful MS/Extra'er
Posts: 46
Joined: Wed Mar 24, 2010 10:00 am

Re: dead vnd5n07

Post by EFIDART512 »

I don't use the data sheet circuit. There is a 36V Zener but it doesn't carry the flyback current. And I use the TIP122 which is rated at 100V but they are Darlingtons not FETs. And the VND5N07 is more than just a FET so there are different things that could fail.

Jean
Where does the fly-back go then?
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: dead vnd5n07

Post by racingmini_mtl »

This is irrelevant to this discussion and off-topic.
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
wes kiser
Super MS/Extra'er
Posts: 1402
Joined: Tue Jan 03, 2006 2:49 pm
Location: Charlotte, NC

Re: dead vnd5n07

Post by wes kiser »

I have a microsquirt module based ECU (that Jean designed the boards for), that much be able to somewhat drive low-z injectors (2 per channel/4 cylinder). The way I handled this was a pair of 3.2 ohm resistors the end user could short with jumpers if they had high Z injectors.

There are likely 30 or so of these running "in the field," and we have had one VND failure.

Just wanted to throw another data point in there (a pair of low z injectors with a small series resistance). This user had "stock" injectors (only 400 cc) which were guaranteed to be very near 100% duty, although the car was running for months. This particular driver was visibly burned/damaged, to the point of actually damaging the trace on the module. While external wiring issues would seam a possibility, even with an external dead short there should be less than 5 amps. I have had others that were "DOA" (would not function on stim when new).
86 Rx-7, swapped to 2.3 ford turbo (BW EFR 6758), ms3/ms3x sequential fuel /waste spark, ls2 coils
88 Tbird 2.3t, Microsquirt Module (PIMP), TFI ignition
Metaworld
Helpful MS/Extra'er
Posts: 44
Joined: Thu May 15, 2008 12:17 pm

Re: dead vnd5n07

Post by Metaworld »

I´ve had the same problems with the VND. My solution was du do an external flyback with a z-diode and an 1n4001 connected in series. The banded side from the 1n4001 to 12volt and the banded from the z diode do the vnd. So its short protected by battery reverse. I could lower take off inductiviti with this little trick to 42 Volt. So from now then all is fine for the VND and work for round about 10000 miles.

Kind Regards, Stefan
dontz125
Super MS/Extra'er
Posts: 4200
Joined: Mon May 11, 2009 7:14 pm
Location: York, ON
Contact:

Re: dead vnd5n07

Post by dontz125 »

Which VND pin are you connecting the diodes to? What voltage Zener are you using?
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module
TTR Ignition Systems
EFIDART512
Helpful MS/Extra'er
Posts: 46
Joined: Wed Mar 24, 2010 10:00 am

Re: dead vnd5n07

Post by EFIDART512 »

I decided to use different FETs (12mohm @ 60A, very thermally efficient LFPAK package) with additional external 5 watt 48V TVS diodes for the fly-back. I'm also adding a watch-dog circuit to monitor all the fly-back pulses which will detect open injectors or shorted FETs. Probably way over-kill but I already have to re-ring my expensive motor for a sinking float. I don't want to do that again. The VNDs should be fine with all of the protection they afford but with all of these failures, I don't want to chance it.
If i was to add external protection to the VND setup, i would look for it to clamp at 30V so the VND internal clamp would be bypassed.
dontz125
Super MS/Extra'er
Posts: 4200
Joined: Mon May 11, 2009 7:14 pm
Location: York, ON
Contact:

Re: dead vnd5n07

Post by dontz125 »

I've been doing some quick reading on the LFPAK mosfets you mentioned; while they do seem very well designed for thermal loading and reduced resistance, I noticed a few things that I hope wiser heads can comment on. The first is that none of the pieces I looked at had any sort of voltage or current clamping features; the VND is a fairly complex circuit unto its own, while the LFPAK devices seem to be "dumb" designs - just big lumps of silicon etc. Especially considering that the cause of the failures remains unknown, this may or may not be a selling feature - YMMV. :D

The second point is the "avalanche" rating. The VND has a claimed Eas of 200mJ at 20v, while the NXP LFPAKs I've been looking at range from 70-100mJ (at various voltages). I have only the most basic understanding of avalanche ratings; "more" would seem to be better than "less", but are the devices operating in the same manner, and are the flyback loads experienced in 12v / 12-16 ohm injectors sufficient to bring these characteristics into play?

I noticed that the body diodes of the LFPAKs were all indicated as "diodes", while that of the VND is a Zener diode. At what point does this become significant, if ever?
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module
TTR Ignition Systems
EFIDART512
Helpful MS/Extra'er
Posts: 46
Joined: Wed Mar 24, 2010 10:00 am

Re: dead vnd5n07

Post by EFIDART512 »

dontz125 wrote:I've been doing some quick reading on the LFPAK mosfets you mentioned; while they do seem very well designed for thermal loading and reduced resistance, I noticed a few things that I hope wiser heads can comment on. The first is that none of the pieces I looked at had any sort of voltage or current clamping features; the VND is a fairly complex circuit unto its own, while the LFPAK devices seem to be "dumb" designs - just big lumps of silicon etc. Especially considering that the cause of the failures remains unknown, this may or may not be a selling feature - YMMV. :D

The second point is the "avalanche" rating. The VND has a claimed Eas of 200mJ at 20v, while the NXP LFPAKs I've been looking at range from 70-100mJ (at various voltages). I have only the most basic understanding of avalanche ratings; "more" would seem to be better than "less", but are the devices operating in the same manner, and are the flyback loads experienced in 12v / 12-16 ohm injectors sufficient to bring these characteristics into play?

I noticed that the body diodes of the LFPAKs were all indicated as "diodes", while that of the VND is a Zener diode. At what point does this become significant, if ever?
You are correct, it is a dumb device.
The avalanche is a condition when the drain-source voltage exceeds the bulk break down of the MOSFET, adding an external clamp handle the fly-back takes this condition away from the device. There is no strain on the device with an external diode/TVS/Zener to clamp that energy. Now the external diode must be able to handle that of course.
The MTD3055VL that Saab uses as mentioned by Keithg is rated at 72mJ (Vdd=25V Id=12A) and Fairchild states that the device has a rugged diode internal diode that can eliminate the need for an additional Zener.
Must be good enough, he says they rarely fail.
The IR part (IPS1041R) that is similar to the VND has an Id de-rating for inductive loads. At 7mH inductance the max Id current approaches 1A. Now this part is not an exact copy of the VND device but I am sure that the VND should be similarly be de-rated for high inductive loads also.

I am not trashing anyone for their choice with the VND5N07 part. I probably would have picked the same part, internal clamp, thermal protection, etc... It looks good.
I will get some of those FETs next week and try to drive a really big inductive load with the external clamp.
Reliability and fail safe is important to me.
dontz125
Super MS/Extra'er
Posts: 4200
Joined: Mon May 11, 2009 7:14 pm
Location: York, ON
Contact:

Re: dead vnd5n07

Post by dontz125 »

What would the clamp consist of - simply a 1N4007 pointing from the source to the drain? Or combined with a 30-ish-volt Zener fitted opposite to the diode?
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module
TTR Ignition Systems
SymTech Laboratories
Super MS/Extra'er
Posts: 2188
Joined: Sun Aug 31, 2008 4:02 pm
Location: South Florida, USA
Contact:

Re: dead vnd5n07

Post by SymTech Laboratories »

dontz125 wrote:What would the clamp consist of - simply a 1N4007 pointing from the source to the drain? Or combined with a 30-ish-volt Zener fitted opposite to the diode?
To clamp the reverse EMF spike to ~12V, install a 1N400x diode between the VND5N07's drain and +12V, with the diode's cathode connected to +12V. To clamp it to a higher voltage, install a 3W (or larger) Zener diode between the VND5N07's drain and ground, with the diode's anode connected to ground. You can add a forward biased 1N400x diode in series with the Zener diode to make a makeshift TVS and protect against negative voltages.
SymTech Laboratories, LLC ----- MSQ File Repository: megasquirt.symtechlabs.com
Image
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: dead vnd5n07

Post by racingmini_mtl »

Clamping to 12V will make for a very slow closing time.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Post Reply