Smoked Capacity Tester

[robbob2112]

Seems like time to buy a higher powered transistor and use the heatsink compound...

 

[Wolfpack7483]

I use the same load tester, and have blown up two of the semiconductors under the heat sink. It’s pretty startling after the first.

Despite the, in my case, the 150 watt rating, it turns out that at higher voltages like 24 and 48 the combined dissipation and switching voltage derate the max load spec significantly.

At 48 volts I keep the power below 50 watts.

The transistor is replaceable and I have two more spares for that purpose.

Thanks. Do you have a link for the transistors? I'd like to try & fix mine if I could.

 

[efficientPV]

You need to put about three more FET in parallel with that one to do any power. I have a 150W load tester and it is five times parts than that one.

 

[Wolfpack7483]

You need to put about three more FET in parallel with that one to do any power. I have a 150W load tester and it is five times parts than that one.

That's probably why it's only $43 on Amazon.
https://www.amazon.com/dp/B07HB3XK7L/?psc=1

 

[robbob2112]

I would desolder the transistor that is there and read the number off the other side of it to verify they match what @effihas... bonus if you lookup the datasheet and get the same spec, but higher powered version....

 

[Wolfpack7483]

I would desolder the transistor that is there and read the number off the other side of it to verify they match what @effihas... bonus if you lookup the datasheet and get the same spec, but higher powered version....

I took it off the board & figured out it is an IRFP264N; TO-218. Looks like it's rated for 250v, 280w max. I looked for similar ones & a 260N; TO-247 is rated 200v but 300w. A 460 is rated to 500v, but still only 280w max. Which would you recommend?

 

[robbob2112]

hrm, not much selection that just leaves having a better thermal bond to the heatsink...

 

[Wolfpack7483]

hrm, not much selection that just leaves having a better thermal bond to the heatsink...

At least I've got good paste!

 

[robbob2112]

I suppose you could attach a water cooler instead of a simple heatsink and get some more power out of it... that way you could burn other bits of the board.

 

[robbob2112]

HEY - here is an idea - run it UNDER water

Sometimes I am just a damned genius

 

[wpns]

HEY - here is an idea - run it UNDER water

Sometimes I am just a damned genius

DI (deionized) water isn’t conductive, we used to run laser tubes in DI water for cooling and they had some serious voltages on them.

 

[robbob2112]

NO, I should have said salt water --denser so it can hold more heat

 

[wpns]

NO, I should have said salt water --denser so it can hold more heat

How about heavy water?

 

[robbob2112]

I didn't say I wanted to glow in the dark... and that might throw off the capacity readings

 

[efficientPV]

You don't need a bigger FET, you need more of them.

 

[robbob2112]

You don't need a bigger FET, you need more of them.

Use some small solid 18 or 20awg wires like from rs485 cables and solder up another one in parallel to the first ... just off to the side maybe a heatsink pad under it them grease on both...

 

[wpns]

You don't need a bigger FET, you need more of them.

And either way, you may need more drive power.

 

[robbob2112]

Hrm,

Use the power to cool itself

https://www.amazon.com/dp/B0C7QHZMG7/

 

[Wolfpack7483]

Use some small solid 18 or 20awg wires like from rs485 cables and solder up another one in parallel to the first ... just off to the side maybe a heatsink pad under it them grease on both...

I've ordered the FET to replace the smoked one, but also saw this in another thread, and it looks a lot more robust than mine that failed, & not that much more expensive - ordered the 450w.

https://www.aliexpress.us/item/3256803271477363.html?gatewayAdapt=glo2usa

 

[robbob2112]

https://www.amazon.com/dp/B0BK8YB2F4?

This is mine - the one you linked is higher capacity but only if you have 4 of them linked together....

With the single it only goes up to 11amps discharge.... still a transistor under the heatsink... I have seen something where someone replaced the 4 resistors with much larger wattage ones and turned it up higher .... They do get super hot after running a while

 

[DVD]

I still use it but these are just poorly designed. IIRC the transistor is mounted print side against the heat sink for one (I’ll stress the IIRC because I don’t retain memories too well).

 

[Wolfpack7483]

I still use it but these are just poorly designed. IIRC the transistor is mounted print side against the heat sink for one (I’ll stress the IIRC because I don’t retain memories too well).

The print side of mine that failed, was not against the heat sink. The opposite side was.

 

[efficientPV]

You just can't get heat away fast enough from a FETs junction due to thermal resistance regardless of the power rating. This is why multiple FET are used. Extra FET can be added in parallel by just mating pin to pin usually. The source pin on the right should get special treatment. Add a foot of small wire like #20 to add some extra resistance. Thos will balance the currents among the FET so they share more evenly. Mount FET directly to heat sink without insulator.

 

[Zwy]

The only adjustments are for current - Coarse & fine. As you adjust, it indicates the power in the readout. Instructions say to rotate both knobs fully counterclockwise (to zero) before test, which I did. I'd planned on about a day & a half for the test.

I had one like that a few years ago and the knobs were switched in the instructions. There is a coarse knob and a fine knob and that is what caused the problem.

After that episode, I just built a tester using a shunt, inverter that powers incandescent light bulbs and a blower fan to increase the load. Light bulb utilizes an old style dimmer that I bought a few on clearance for 2 bucks each.

 

[robbob2112]

For the most recent 280ah 12v battery I hooked it up to an inverter with a know efficiency rating ... then hooked up a 1500w ceramic heater through a Victron shunt... It was so much faster to do the test even if the results are a few Whr one way or the other, it doesn't really matter to me.

I will only be using this small capacity tester to run the individual 314ah cells at 3.2v