Trident -- Unusually Loud Fan, Bad Tests, flickering status light - Problems and Solution

[gbru316]

Earlier today, I noticed my Trident making a ton of noise. Not the normal whizzing, clicking, and whirring. It sounded like a miniature jet had located itself inside the unit. Plus, the status light was flickering while the Trident was sampling/testing.

I did some digging and determined that the noise was coming from the fan underneath the cuvette. It was spinning at max RPM. Not only while testing, but also while idle. This caused a secondary problem in that the fan was spinning too fast for the magnetic stirring rod to remain coupled with the fan so it was just kind of lazily bouncing around in the cuvette. In other words, it wasn't properly mixing the reagents and sample causing test errors.

I opened up a support ticket but then my curiosity got the best of me. This is what I found:

It seems, somehow, that sample/reagent leaked from the cuvette and spread over to the microcontroller PCB -- specifically to the power/fan headers. Once there, it caused significant conductive corrosion between the header pins and, if I had to guess, shorted the fan pin PWM pin (yellow) to the fan DC source pin (red). The flickering status light can be explained by a conductive short -- (or at least, sufficiently changing trace impedance if not a full short) -- between the blue microcontroller communication wire and the fan PWM wire. Again, guessing here, but it's probably either I2C or SMbus used to communicate between Apex and the ATTiny416 uController (located on other side of board).

I removed the corrosion with some isopropyl alcohol, verified my cuvette wasn't cracked, reassembled, reinstalled, and tested the Trident. Surprisingly, it works just fine now with the exception of needing a re-cal. Easy enough fix. Just in time for my shipment to arrive tomorrow for the Battlecoral Grow-Out Competition!


The question remains, though -- how did liquid escape the cuvette? The tube looked secure. The cuvette isn't cracked. I sure didn't move the Trident while it was testing. The cuvette lid is in-place and secured with tape (from the factory). I'm genuinely stumped. Either the tube isn't as secure as it looks, or some SW/FW bug allows the cuvette to be overfilled in certain conditions (I read of at least 1 beta tester having an overfill condition).

 

[SPR1968]

I can’t answer with the question but I can bump the thread to see if we can get you some help

 

[gbru316]

I can’t answer with the question but I can bump the thread to see if we can get you some help

seems like there is, indeed, a condition that can cause the cuvette to overflow. It was reported several times during beta testing and when I spoke with Neptune about this they didn’t seem terribly surprised.

they’re cross shipping a replacement which is supposed to arrive Thursday.

i just wanted to document this in case this happens to others so they can (hopefully) get their trident operational while they wait for the replacement to arrive.

 

[Jonny Walker]

This exact thing is happening to me as I write this
I'm going to check if any liquid spilled in mine

 

[Jonny Walker]

This exact thing is happening to me as I write this
I'm going to check if any liquid spilled in mine

Well well well...is there anyway way to prevent this?

 

[Simon Reefing]

Following

 

[Gtinnel]

Well well well...is there anyway way to prevent this?

I don’t know what causes it to overfill but the same thing happened to me about 6 months ago. I tried to get them to sell me just the little board but they refused, so I paid for the advanced replacement maintenance.
I’m determined if it happens to me again I’m going to try to coat the board with something to try and make it water resistant.

 

[vetteguy53081]

Earlier today, I noticed my Trident making a ton of noise. Not the normal whizzing, clicking, and whirring. It sounded like a miniature jet had located itself inside the unit. Plus, the status light was flickering while the Trident was sampling/testing.

I did some digging and determined that the noise was coming from the fan underneath the cuvette. It was spinning at max RPM. Not only while testing, but also while idle. This caused a secondary problem in that the fan was spinning too fast for the magnetic stirring rod to remain coupled with the fan so it was just kind of lazily bouncing around in the cuvette. In other words, it wasn't properly mixing the reagents and sample causing test errors.

I opened up a support ticket but then my curiosity got the best of me. This is what I found:

It seems, somehow, that sample/reagent leaked from the cuvette and spread over to the microcontroller PCB -- specifically to the power/fan headers. Once there, it caused significant conductive corrosion between the header pins and, if I had to guess, shorted the fan pin PWM pin (yellow) to the fan DC source pin (red). The flickering status light can be explained by a conductive short -- (or at least, sufficiently changing trace impedance if not a full short) -- between the blue microcontroller communication wire and the fan PWM wire. Again, guessing here, but it's probably either I2C or SMbus used to communicate between Apex and the ATTiny416 uController (located on other side of board).

I removed the corrosion with some isopropyl alcohol, verified my cuvette wasn't cracked, reassembled, reinstalled, and tested the Trident. Surprisingly, it works just fine now with the exception of needing a re-cal. Easy enough fix. Just in time for my shipment to arrive tomorrow for the Battlecoral Grow-Out Competition!


The question remains, though -- how did liquid escape the cuvette? The tube looked secure. The cuvette isn't cracked. I sure didn't move the Trident while it was testing. The cuvette lid is in-place and secured with tape (from the factory). I'm genuinely stumped. Either the tube isn't as secure as it looks, or some SW/FW bug allows the cuvette to be overfilled in certain conditions (I read of at least 1 beta tester having an overfill condition).

It got wet and likely from leaking or moisture. Cap appears to be in place on the cuvette. Overfilling can occur if manifold is clogged somewhere internally
Ticket was good as they will instruct you to send it in for further assessment but this will likely void any warranty.
You may get lucky and they will offer 1/2 price on a refurb unit or exchange

 

[Gtinnel]

You may get lucky and they will offer 1/2 price on a refurb unit or exchange

If the trident is out of warranty they will give the option to do the any of the trident service programs (excluding the diy) to get it fixed. With the trident service they will fix any issues with a trident so at least it only takes $200 to get it fixed.

 

[Jonny Walker]

I cleaned it with isopropyl and blew out the lines with a keyboard air can
It seems to be working now.
I am going to try to do some more research on why this would happen to try to prevent it in the future.

 

[Gtinnel]

I cleaned it with isopropyl and blew out the lines with a keyboard air can
It seems to be working now.
I am going to try to do some more research on why this would happen to try to prevent it in the future.

If you figure something out please don’t hesitate to share it.

 

[Fearviper]

Take the board out....clean it with alcohol...put the board back in, where you seen the corrosion, apply a thin coat of silicone to water proof it

 

[stoney7713]

I don’t know what causes it to overfill but the same thing happened to me about 6 months ago. I tried to get them to sell me just the little board but they refused, so I paid for the advanced replacement maintenance.
I’m determined if it happens to me again I’m going to try to coat the board with something to try and make it water resistant.

Ha! I know of something! Corrosion X! I used it when I flew RC planes. Especially useful for those pesky seaplanes when they wanted to take a bath.

Just let it dry and it works really well. There's so many uses for it, you'll find yourself reaching for it all the time. I still use it though I don't fly anymore.

 

[Gtinnel]

Ha! I know of something! Corrosion X! I used it when I flew RC planes. Especially useful for those pesky seaplanes when they wanted to take a bath.

Just let it dry and it works really well. There's so many uses for it, you'll find yourself reaching for it all the time. I still use it though I don't fly anymore.

I’ve never heard of it, but is it a spray on coating that stays on the board to protect it? From reading a quick description it seems like it’s more to spray on after there is already corossion.

I’ve never used it but I was looking at this product Which seems to be a spray on rubber coating. The only reason I haven’t done it is because the board I want to coat has the photocell on it and I assume I can’t spray the sealer on the cell even if it is clear, and I’m a little nervous to mask it off and spray the rest of the board.

It’s a shame the Neptune doesn’t coat these boards or make them potted from factory since apparently water spilling inside the unit is not uncommon.

 

[stoney7713]

I’ve never heard of it, but is it a spray on coating that stays on the board to protect it? From reading a quick description it seems like it’s more to spray on after there is already corossion.

I’ve never used it but I was looking at this product Which seems to be a spray on rubber coating. The only reason I haven’t done it is because the board I want to coat has the photocell on it and I assume I can’t spray the sealer on the cell even if it is clear, and I’m a little nervous to mask it off and spray the rest of the board.

It’s a shame the Neptune doesn’t coat these boards or make them potted from factory since apparently water spilling inside the unit is not uncommon.

Yes! It removes rust and corrosion and forms a molecular bond one the outside of the metals. It prevents rust and corrosion from happening again and can make circuit boards waterproof.

We would dip our boards in it to make sure it coats everything and let them dry. Never ever had a short again after using it.

I would avoid the rubber coatings and such, they trap and retain heat, which can cause failures. Not to mention once it goes on it doesn't come off.

It's highly recommended for cleaning guns, waterproofing boating equipment, especially saltwater crafts, aircraft, farm equipment. I really don't know why it hasn't caught on more, other than it is a bit expensive. But a little goes a long way.

There's a few different types but they are all about the same, I just use the original squirt bottle. Been using it for over 2 decades. It's not like WD-40...

Read up on their website how it works. I just wouldn't use it on anything we submerge in our tanks

CorrosionX – Corrosion Technologies

Kills Rust and Corrosion World's Greatest Lubricant, Penetrant & Anti-Seize Safe on Electronics CorrosionX is the most advanced and effective corrosion prevention compound, lubricant and penetrant in the world! Although it may look like a typical oil-based anti-corrosion spray, CorrosionX...

www.corrosionx.com

Edit: the photo cell would be the only thing I wouldn't want to get it on. I don't know how it would affect it. Maybe take a small brush and brush it on the other parts of the board.

 

[Coolcasino]

Not ment to highjack but since some of you have experience with this... Mine crapped out on me today. The mixing fan wont turn on. I could see the pill not spinning.. Thought it might be corrosion or the board getting wet somehow but my board is clean. Thought the fan was the issue but I can turn the fan on when I hook it up to a power supply. So I guess the trident isn't able to power the fan for some reason. Got the automated email from Neptune stating that according to all the troubleshooting I did it will have to be sent in for repair. Has anyone had this happen the them? Any ideas as to why the trident isn't powering on the fan?