error fpga fan speed

Gene - K3GC

I would love to know what the failure rate is for these fans.  I suspect that it is quite low.  Satisfied customers with no problems are not likely to post to a forum such as this or even read it for that matter.

If FRS were to post the actual failure rate for these fans I suspect it would alleviate concern for all who are concerned about possible failure of fans.

Thanks

Delbert McCord

The fact that they have not to me says they do have a problem.

Bill -VA3WTB

I think they have, Steve said above they discussed the fan failures in a meeting. His reply was very good and honest.

For me, I know some have had problems, but I never think about,,I just use my radio the same every day, I feel the probability of a fan failure is remote. But possible like anything else.

Harold Rosee

Steve,

To use your analogy when a vehicle has a problem the vehicle is recalled and fixed with a different part or firmware. It is usually covered if it's out of warranty.

It appears that the repair FRS is performing is o install the exact component that was the problem in the first place.  I see that several have failed more than once.

I would think that FRS would come up with a fix that would use a different fan even it it required a firmware change that worked for either fan.  To me, it is unacceptable to fix them like they are currently being repaired.  I mean no disrespect but it makes no sense at all to me to do that.

I have not had this problem with my 6500 but I don't operate that often.  Mine is still in warranty but it sure give me pause as to whether to keep it or not.

How about a better solution.  I think that is really the point of all this discussion.  Even if we have to pay for an out of warranty repair it would be easier to swallow if we knew if was a proper fix.

Thanks for your consideration and comments on this.

Harold
W5ZZT

Douglas Maxwell

It would be cool if this was a passive fix with a falsified fan rev input mod.

Gene - K3GC

Failure of a part per se  does not mean that the part is overly prone to failure.  If they fail at a rate of 5/thousand per example that would most likely be the luck of the draw. Moving parts WILL eventually fail. If the failure rate was 50/thousand then a new part may be in order.

The recall analogy is that a general recall is done due to either safety concerns or a part that fails way too often.  

We do not have that information.  The fact that a large number have posted to this forum  does not really tell us anything except that people with problems are most likely to complain. I suspect that there is a very large number of satisfied Flex 6xxx owners out there who do not even use this forum.

If I had the data that the failure rate of these fans was very low I could go about happily using my radio without another thought.  If, on the other hand, the failure rate was high I would want to have a couple of fans in reserve.

Steve's letter does not deal with actual failure rates and that is what I would like to see.

Bill -VA3WTB

I think you mentioned that before Gene

Gene - K3GC

Just a bit more specific

Steve-N5AC

I do not know the current failure rate in terms of failures/radio.  We've not taken the time to go compute that number yet.  This came up for discussion because it rose to the most common failure (outside of lightning) in the service shop.  We keep an eye on things like this and if it seems odd, we go investigate if there is an issue.

Someone from FlexRadio will post the part number next week when we get a minute to do so.  I don't have it handy this weekend.

We always look at any failure with a mindset of questioning the root cause of the failure and discussing if it is a design issue, a supplier issue, etc.  If it is a supplier issue, we address these directly with the supplier and generally arrive at a joint resolution that works for us, our customers and the supplier.  In the event of a design issue, we decide how serious the issue is and if it warrants a change in the design, a retrofit, etc.  We've done a few retrofits after ship in the field for issues we considered significant.  We're still "studying" the recent uptick in fan failures.  In terms of raw percentages, I believe the numbers are fairly low, but again we've not computed them yet.  After doing some analysis on the "why," we will decide if further action is warranted.  

Al_NN4ZZ

Hi Steve,
On the most common repair issue ( Lightning damage) can you share what are the common types of part failures you see?  

I'm curious if you have seen any issue with the radio USB ports.

If the radio is off (powered down) and the antennas are disconnected then I would suspect the ethernet port would be most likely.  It would be interesting f you could share any numbers of the types of failures.  

https://community.flexradio.com/flexradio/topics/dead-ethernet-port

I addressed the ethernet port with a fiber link a few years ago and I disconnect the antennas and power when it is not in use but recently someone asked me about the USB ports.   My current thinking is that powering all the USB devices from a UPS (to address power line surges) and using a (sacrificial) USB hub between the devices and the radio USB port is adequate.  

Thoughts? 

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
SSDR / DAX / CAT/ 6700 -  V 1.10.16
Win10
  

Gene - K3GC

Thanks Steve,
Your response is most helpful.  I will look forward to the followup 
Gene K3GC

Manuel - W4SSB

Did the part # for the fan ever get posted?

Gene - K3GC

Not yet

Steve-N5AC

The part number for the fan we're using is AP0512HB-J96(S)-4P-TA-LF.  This part is at end-of-life (we've used it for about 6-years now) although we have stock.  We have an engineering project to investigate a suitable replacement fan or other technology to keep the parts cool for the future, but for us this is not urgent since we have replacement parts.  As you might imagine, parts go EOL on a regular basis and part of sustaining engineering is to find replacement parts that work in the design so this is "par for the course" as an electronics company.

The fan is currently attached to two parts: the main CPU and the FPGA.  We have two mechanisms for ensuring that we are in temp specs.  We have a tachometer on the fans (both are monitored and the radio will remove power from the chips if the fans stop) and the FPGA has an internal temperature sensor.  With the FPGA we provide a warning if it's getting too hot and an error/shutdown if it gets to the temp that we want to stop operation.

FYI For the newest radios, the FLEX-6400 and FLEX-6600, we've moved to a single, larger fan and mechanicals to direct the airflow across the proper parts to reduce the number of moving parts in the radio.  So neither of these radios happen to use these smaller device fans.  This doesn't alter our need to look for a longer term solution for replacement parts for the 6300/6500 and 6700, of course, but is just provided to answer the curiosity question should it arise.

k0eoo

@Al,

On your USB port question regarding lightning issues, about 3-4 years ago I experienced a hit to the tower while all my antenna's and ports were connected.  The 6500, PC, rotor controller, DX engineering Remote ant switch, Steppir controller, LP-100A and Alpha 87A were all connected to the PC via hub's and serial to USB adapters.  All the gear was turned off.  

After the lightning strike, ALL the serial to USB adapters, hub's and PC were blown.  Anything connected to the USB was seriously effected.  The LP-100A, rotor controller and 6500 were not effected.  The DX engineering remote control switch box was blown as was the SteppIR control box.  The RS232 chip in the Alpha 87A was blown as well.

I guess this is the long way around to my point, and that is, if all my USB stuff had been connected to the 6500 instead of the PC I think I would have probably blown the 6500.  Since that incident I have added an isolated USB Hub that provides about 4kv of isolation to the PC.  If you're interested in the model number let me know.  

I was out of town when that lighting incident occurred and unable to disconnect everything like I usually do.

Douglas Maxwell

Could Flex also consider a mechanical clip to stop the heatsink and fan assembly becoming unstuck from the chip due to its own weight and it's inverted orientation. This would increase the possibility of me taking my flexradios out portable.

Al_NN4ZZ

@Dennis,
Thanks, here is the USB hub I'm using, it's advertised as having "industry level protection."   Not sure how that compares to the one you are using.  What model are you using and do how do you think it compares?

I'd rather replace the hub than send in the radio since it is only $26 from Amazon.  I also have the hub power supply and all the USB connected devices connected to a UPS for power surge protection.  

I asked @Steve H for feedback since I was curious if FRS had any other recommendations for USB protection and whether they have seen any USB failures.  In another related thread Tim indicated they weren't going to reply about the lightning failures other than to say that the ESD diodes were the most common one. 



Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
SSDR / DAX / CAT/ 6700 -  V 1.10.16
Win10

k0eoo

@Al,

Here is the one I bought, UHR304.  http://www.bb-elec.com/Products/Datasheets/UHRx04-2512ds-pdf.pdf

Its a USB port ISOLATOR.  The PC provides power for the HUB side and a separate well isolated power supply provides power for the USB inputs.  In between it provides 4kv of isolation and 15kv of ESD protection.  It was a bit pricy but USB port isolators that will work at FS are hard to come by, at least 3-4 years ago they were.  The one you show does not say it provides ISOLATION between the PC and the USB port inputs so I'm not sure it will protect your rig.

I might add, if you get a direct lightning strike and all your ports are connected 4kv of isolation won't help, this will only help protect in static discharge situations, at least I hope it does.  So far I have had NO port issues since installing the USB isolator, knock on wood....

Delbert McCord

Ok so Flex has some fans still is stock. Can I buy some? That was one of the original questions knowing that they are discontinued and I have 2 6300's I do not want to be stuck with expensive paperweights in a couple years.

Bill -VA3WTB

I think Steve made it clear that Flex will not let the radio turn into a paper wight. He said they are looking to other options for cooling,,  So I don't see the panic here.

Delbert McCord

I am not panicking I just want to have some spares. I already had to send one of them in for the fan problem and I have a friend that is about to send his in for the second time.

Al_NN4ZZ

Dennis / K0EOO,
Agree the hub I have advertises "industry level protection" but doesn't say what that means and I don't see any specs on isolation numbers on their web page.   My assumption was that it would be better than directly plugging into the radio USB ports and act like a $25 fuse.  

The hub you got is clearly better and looks like a great USB 2.0 hub but it sure is expensive at about $300.  They also make a 3.0 USB version for $375.  

Are you using yours with the Flex radio USB ports?   What devices do you have attached?   I'd want to be sure there are no glitches before spending that much.








Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
SSDR / DAX / CAT/ 6700 -  V 1.10.16
Win10

Al_NN4ZZ

@Don / W2XB
I hope you don't mind the discussion about the USB ports here.  I know this was intended to be a post about the fan failures.   It started when Steve mentioned that the fan failures were second only to lightning failures.   Anyway I should probably have copied that comment and started a separate thread.   I originally thought it would result in a quick answer but it evolved to much more.  

So I apologize for that....

Regards, Al / NN4ZZ  
al (at) nn4zz (dot) com
SSDR / DAX / CAT/ 6700 -  V 1.10.16
Win10
 

G8ZPX

interesting sentiment and anecdotes from N5AC, but lacking any facts about reliability, MTBF figures, part numbers, why they are failing so early and repeatedly, what is valid technical solution for the end-of life-part code in a raido only 3 years old?

Did I say part number?

Gene - K3GC

I contacted Mouser last week regarding a replacement for the obsolete fan.  They sent me data sheet for the replacement.  They have to be ordered in lots of 20 at $13.12 ea.
The first thing I noticed was that the fan only has 2 wires (no tach)
For those of us who would like to have spare for later self install the only option appears to be Flex unless someone can find another source.

k0eoo

@ Al,

Hi Al, my UHR304 USB port isolator has been connected to my PC for around 3 years now.  NO issues with it hurting the PC or anything.  When the day comes that I switch over to using the Flex USB ports I'll use that same HUB.  As far as issues go, once or twice a year I have to reset the HUB by powering it down for a couple seconds, it always comes right back up and all works great!  Normally I leave it powered up 24/7.

As I mentioned before, because USB data streams are not symmetrical, they cannot easily be isolated, like with a capacitor or transformer.   Consequently, there is no cheap way, I know of, to isolating USB data making the HUB's quite expensive as you've mentioned.  For example, Ethernet ports on everything are DC isolated to about 1500v because Ethernet data protocol, simulator to 8bit/10bit coding, is symmetrical and can be easily isolated

You mention a $25 fuse, in my previous experience the old cheap HUB I had passed on the lightning surge and took out the PC USB ports and PC, it did not protect the PC so I expect it wouldn't protect the Flex.  Part of the problem might be how you power the HUB.  If you power the HUB from the PC like I was doing any glitch on the ports would pass right on to the PC, but, if you use a separate PS for the HUB like Flex recommends, that might add an unquantifiable level of protection?

I don't know Al, do you think we have beat this dead horse long enough, LoL?

Tim - W4TME

2-wire fans will not work.  The radio will not boot without the tach line.

K2CB Eric Dobrowansky

See G7BCX's post a week ago in this thread. There is a fully compatible replacement. Same rpm monitoring function. Plug an Play, just that it uses a newer "hyper bearing" internal design. PN AP0512HX-J96

They are not normally stocked by the distributors (Mouser, DigiKey, etc). 

As I mentioned earlier in the week, if there is enough interest, I can do a bulk (100pc) purchase direct from the manufacturer. Price will be just below $15 each, plus postage.  

Eric
K2CB

Gene - K3GC

The data sheet I got from Mouser does not show the 
Tach line which as Tim said will not work.
Link to data sheet: http://www.mouser.com/ds/2/6/adda_ap5008-608356.pdf

K2CB Eric Dobrowansky

That's because that is not the correct replacement.  The PDF referenced above is for the J90 version which is a two wire design. The J96 version has the third-wire rpm monitoring circuit.

Here is the correct version. 

http://www.mouser.com/ds/2/6/Adda_AP0512HX-J96-A-347536.pdf