Partial Jag failure?

[kamocat]

Well, it worked fine today. I even held it in a 41 Amp stall for 5 seconds.
There's only two causes I can think of.

One is that perhaps it had tripped the breaker and needed "voltage enable" to be sent again. However, I'm pretty sure I would have seen movement out of it before that happened.

The other is that today I'm outside at 15 degrees Celcius. Friday I was inside at 30 degrees. The Jaguar only said its internal temperature was 37 degrees on Friday. I thought it would operate up to 50.

(Note: I haven't opened it up yet.)

[EricVanWyk]

Marshall - does it respond to PWM normally? Are there any signs of physical damage?

Originally Posted by Al Skierkiewicz

...

Al -

Thank you for the very clear and informative post.

The difference in total on-resistance between the Jaguar and the Victor is inconsequentially small. I think the Jag wins the numbers game, but not by enough to make an actual difference.

The reason that the Jaguar uses fewer FETs is that it uses slightly more modern FETs that can handle more current per device.

[diviney]

Originally Posted by Mike Betts

JW,

If you look at the spec sheet for the MX5-A40, you can pull 133A through the circuit breaker for about 3 seconds before it trips. This can be a long time for a resistor to gain heat depending on the thermal resistance of the resistor.

Mike:

I was looking for breaker information when I found your post... You state that the MX5-A40 can deliver 133A for around 3 seconds. As I read the data sheet, 133A represents around a 300% overload, which would trip in substantially less than 3 seconds (0.5 to 1.1 seconds). Am I missing something?

-Tom

[Al Skierkiewicz]

Tom,
You reading is correct. The smaller breakers have a slightly different trip spec and the 120 amp breaker is longer still. However, all of the resettable breakers will have a shorter trip spec after the first reset as the breaker is already at an elevated temperature. This is also compromised if the breaker is in a hot environment. i.e. near a hard working motor(s) or thermal shadow, adjacent breakers are also passing a lot of current or if the wiring attached to the PD has not been properly terminated (high resistance) and is creating heat from the termination.

[techhelpbb]

Originally Posted by Al Skierkiewicz

Dude,
This is not always the failure mode on the Tan or Black Jag. Your calculations on the current sense resistor are mostly correct and it is likely the source of the failure if nothing else went sour at the same time. I have noted that many tan Jag failures occurred when the current sense became so hot, it unsoldered itself from the board and in the movement of the robot actually moved off it's circuit pads. This opens the current flow inside the Jag. Just to be accurate, CIMs draw 133 amps in stall which would dissipate about 17.6 watts in the current sense resistor. Over an extended amount of time, this would cause severe heating of the resistor. Normally, CIMs are run at 40 amps or less so normal dissipation is about 1.6 watts.

The current shunts in analog meters (and many digital meters with current measuring) is to prevent damage to the meter (either causing the pointer to wrap itself around the pins or damaging analog to digital convertors). The more sensitive an analog meter, the less current it would need to go full scale. I have an old Midland analog meter that has a full scale current of 60 microamps. In order to measure current the meter needed a current shunt across the terminals to prevent damage. Say you wanted to measure 10 amps full scale, then the shunt had to be sized such that 9.99994 amps flowed in the shunt. This shunt is not the same as the series current sense resistor used in the Jags. My Fluke uses a #10 wire as a shunt on the high current (10 amp) range.

I will not disagree with you about your experience in the matter with the analog meters (you are not incorrect about it, I am merely trying to use the terms in this way because I had hoped they would be most clear in this manner). However, different strokes for different folks.

Texas Instruments, the designer of the part INA193, specifically calls it a 'current shunt monitor':

http://focus.ti.com/lit/ds/symlink/ina193.pdf

I specifically provided an official application note above in which those parts are used in a replacement function over a current sensing application...and in the last post of the series...openly noted that it could also be called a 'current sense monitor' or 'current sense amplifier'. In point of fact, other manufacturers do...indeed...use the alternative naming.

For example:

Here national semiconductor refers to the same basic circuit by the alternative name...
http://www.national.com/analog/ampli...ent_sense_amps

Here WikiPedia furthers calling this a current shunt:

http://en.wikipedia.org/wiki/Shunt_%28electrical%29
"An ammeter shunt allows the measurement of current values too large to be directly measured by a particular ammeter. In this case the shunt, a manganin resistor of accurately known resistance, is placed in series with the load so that all of the current to be measured will flow through it. The voltage drop across the shunt is proportional to the current flowing through it and since its resistance is known, a millivoltmeter connected across the shunt can be scaled to directly display the current value."

Wonderful world of engineering...same idea, different words...and we are all using the same language. Let's try it in Korean or Japanese.

This is merely a matter of difference of terminology. This is why I posted that 3rd post.

Further, you are correct...of course...it's not always the MOSFETs that go.
However, it never hurts to ask because the MOSFETs have even more reasons they can go.