The concern of late has been about acquiring fans and DIY replacement. K2CB has been kind enough acquire fans at a favorable price so acquisition is not a problem. Replacement is the next hurdle.
The concern expressed by Flex re DIY fan replacement is the concern about damaging the chip due to the fact that the adhesive is so STRONG.
The other concern expressed by Steve is difficulty replacing the shroud. It doesn't look too difficult so long as you are extremely careful getting all the tabs aligned properly.
I hope someone (Flex?) will post detailed instructions. Meanwhile my 4 year old 6500 has been been cruising along with no problems but, in an abundance of caution I have ordered 2 spares.
Hopefully, Steve will post a failure rate of these fans so we will have an idea how real the concern about fan failure really is.
Gene and all,
I have done the fan replacement recently (I posted original fan pics). They are not at all difficult to change, you should leave the heatsink in place (no need to disturb the heatsink bonding anyway).
Carefully unscrew the fan subassembly from the heatsink whilst holding the heatsink so that it does not try to rotate, then place the new donor fan to the existing heatsink. Check they spin-up okay, and the radio boots happily with the new fans - job done.
The RF cage is pretty large, but NOT difficult to get back on. It has lots of friction grip fingers, and is hampered some by less than idea positioning of it close to the outer case, but it can be done.
The whole job takes about 30 mins and is easy to do - provided you have good eyes and a steady hand. However, it should only be attempted by those that understand which end of a soldering iron gets hot (!) i.e. if you already do component level work then this is a walk in the park. Not sure why Flex make out its such a big deal really.
The only bad thing is that Flex have failed to take a positive lead with good service - especially when the fan are failing so prematurely. The user base has had to ID the problem, find the part code, find a source, and order parts for themselves. A sad reflection on Flex IMHO.
I would like to add my final thoughts on this cooling fan issue.
So clearly these fan’s are a big problem, Now obviously all moving parts like fans will eventually wear out unfortunately these are failing way too rapidly. I’m sure when the engineers at flex decided to use these fans in the radio design they expected them to last a lot longer than they have something they could not have anticipated since the fans are rated by the manufacturer for a minimum of 50,000 hours before failure.
Now add the fact that the manufacturer of the fans have decided to basically discontinue these fans, thankfully there is a reasonable amount of stock still out there to last for a little while so it’s not that desperate just yet. And better than that, some of the users have gotten together and are ordering direct from the manufacturer in bulk while supplies last to get ahead of the depleting part stock.
The problem is that awesome radios like the 63/65/6700 up to a $7500 radio are potentially being “Bricked” over a $15 part.
IMHO the flaw in the design is relying too much on moving parts like these fan’s, something FRS has learned from and that's why the new 64/6600(M) don't have these fans. Keep in mind that these radios are touted as “Radio Servers” and that now with SmartLink more than ever many users will want to leave on 24/7 to take advantage of the new powerful remote capability.
Where do we go from here?
The primary cooling fans in the radio is a pair of Standard 12vdc PC type 80mm fans that are easy to obtain. However the ones that cool the FPGA and main CPU are very specialized and now hard to get.
So as I have mentioned before I think the best way forward is to design a passive cooling option using a large Heatsink, plus you get the added benefit of reduced dust witch lowers the about of times the user needs to open these radios to clean the insides. Now I know it’s easy for me to sit here and Monday morning quarterback the engineers at FRS but I think this is going to be the best way forward.
The engineering questions to be addressed are as follows:
1. Can a large enough Heatsink be make to safely “passively” cool these chips with a reasonable temperate margin or headroom?
2. Can it be made to fit inside the existing RF cage or will a new cage have to be made?
3. Will this large Heatsink create RF issues inside the cage, such as parasitic’s causing the radio to crash or damage the chips?
4. Can the R&D and part be made at a reasonable cost?
I’m sure I missed some other possible issues but these are some of the basic questions.
Here is another question,
What happens if nothing is done to address this issue, I think this could be a disaster for Flex and it’s reputation. Imagine as more and more of these fan failures continue to pop up over time what this could do to FRS reputation and the confidence that people have in their products. Also how ticked off people that have invested thousands of dollars in these radios find that they are unusable is such a relatively short time after purchase. Just imagine the reviews on EHam, QRZ, YouTube and other websites in the ham radio community. All this while the SDR radio continues grow and to be made by other companies with deeper pockets. This is just the kind of thing the competitors would just love to use against Flex.
I have a good feeling that Flex already knows all of this and will fix this issue, restore confidence with it’s valuable customers and avoid a potential Public relations nightmare.
We’ll just have to wait and see.
Steve said they have had meetings on this and they are looking into why some fans fail and others run for years.
I understand there have been some fan failures but blowing this out of proportion is not helpful.
You are part of a very small group who feel Flex has failed in it's servicing in this respect.
Steve explained why he has taken as long replying to all the questions here.
Guys, if we are going to call Flex out over this, then we should keep our comments fair and factual.
I am still not sure going on our own is the best thing and not working with Flex to find a better ground. Steve and the team have been very busy as he said and Flex does not see this as a pressing problem as yet,,so there is no panic. But we should give Flex a chance to work on this and let them report what they find out and are willing to do.
There's an old movie with Robert Redford, Sneakers, where the antagonist makes the comment that mentioning publicly that a bank is unstable causes a run on the bank and the run on the bank causes it to be unstable -- even if it was fine to start with. Panic begets panic.
Yes, some of the fans are failing earlier than we expected. We believe something that has changed with the supplier as the early fans we purchased are having significantly fewer problems that newer fans. We had an internal discussion about fans long before anyone pointed out publicly that there have been fan failures. We are investigating and working toward a solution. Our ME passed away suddenly this year and we have someone to help us now, but we had a backlog of work for him and this hasn't been addressed yet. We will formulate a solution and we will communicate about it when we have it.
We (FlexRadio) will continue to charge some of the lowest service rates in the industry, treat our customers fairly and do all we can to help them. As we have time, we will continue to answer well-intentioned, sincere questions from you guys -- our loyal customers -- on anything that is of interest to you that remotely concerns our products or ham radio. I appreciate that some of you are concerned because of a component failure. Please believe we're doing all we can to get to the crux of the problem and come up with a solution, but continuing to suggest that we might not do something about it (and what a disaster that would be for FlexRadio) just isn't helping. Someone suggested that we are not "providing facts" on actual MTBF of the fans. As an engineer, if I'm asked to report MTBF you sorta have to know the time that the fans are running. We do not collect information from our customers on how many hours a day their radios are turned on, hence the calculation is difficult. It helps no-one for me to get into a pedantic discussion about calculation of MTBF. It also really doesn't matter -- MTBF is a decision support tool. We've already made the decision that we are not pleased with the fan performance and want a replacement solution. Why would I painstakingly calculate MTBF now? All the answer I need as a businessman is "the reliability has decreased with the newer fans and we would like a replacement solution." Please give us the time to develop a solution. And don't panic.
Is there a test or procedure to verify whether the cage has been replaced correctly? Certainly if you are getting lockups from RF that would be a clue since it's purpose is to keep RF from the receiver.
Would there be any benefit in looking at before and after snapshots of the panadapter to see if there is any significant change in the transmit signature?
Does the FRS service team have anything they can share? I wouldn't be surprised if it is just a re-install carefully and visual check but it doesn't hurt to ask.
Regards, Al / NN4ZZ
al (at) nn4zz (dot) com
SSDR / DAX / CAT/ 6700 - V 1.10.16
Speaking of the shielding cage, if anyone has any photos of the opened cage, original fan, and actual pin connector where it plugs into the board, I would greatly appreciate an email with a photo or two. I want to be 110% sure the mfg is supplying me with the correct connector and lead length, so it is truly a plug and play swap. They have asked me for a photo of the original unit and connector so as to verify.
Also, with the photos, I could possibly photo shop them and prepare some sort of how-to doc to enclose with the fans when they ship.
BTW, we have met the minimum, but there is still time to get on the list, as I can increase the quantity if needed.
But all models have 2 fans inside the cage which are the ones Eric / K2CB is offering. One for the CPU and one for the FPGA. I ordered 2 and figured if I have to remove the cage to replace one, might as well replace them both. (although the only error I've seen reported is the FPGA error so wonder if there is a CPU error as well or do both get reported as the same FPGA error)
@Eric, what about replacements for the other fans? Should there also be a supply of the other one (1) or two (2) main cooling fans at the front of the chassis? Assume they are a different model, is that correct?
I haven't seen much about the failure rate for them or whether they generate any specific error codes. But in a few posts there have been references like "FRS replaced all 4 of my fans."
Regards, Al / NN4ZZ
al (at) nn4zz (dot) com
SSDR / DAX / CAT/ 6700 - V 1.10.16
It is a full drop in replacement. Same OEM manufacturer, same wire color coding and connector, same heat sink, and the same thermal adhesive pad. The only difference is the bearing technology, as the oem no longer offers the older style design. I am told the Hypro bearing is better, but I'm no fan design expert!
Most will opt to just unscrew and swap the fans, but for the purist, this is a complete replacement. No cutting corners. See pics below.
For those who preordered, I expect the fan order to arrive here Wednesday, and they should be in the mail and on the way to you by Friday. Watch for your shipment notification email.
For those who may wish to purchase fans in the future, I will have a limited stock on hand for $20 each, plus $10 flat rate shipping in the US. Inquire for shipment rates outside the US. If you think you want to purchase a fan, don't wait too long, as I don't foresee ordering another 100 fans anytime soon!
Thank you to everyone who preordered and made this possible!
The fans arrived yesterday, and were shipped out to everyone last evening. Everyone should have received their tracking notification emails by now, and those in the US should expect delivery by Saturday. Outside the US, allow a week or two.
Thanks again to everyone who participated and made this "group buy" possible.
For those who missed out, I have approximately 8 fans left at this point in time. $20 each plus $10 shipping. Once they are gone, that's it!
There is a lot of good information in various places about the FAN replacement options. Since it may be helpful to have it all in one place, I updated my web page with pictures, video, and links for the parts. Thanks to Eric, Manuel and Dave for the info. I'll keep adding to this as more becomes available.
Regards, Al / NN4ZZal (at) nn4zz (dot) com
Replacing the fans took about half an hour. The video in this thread was very helpful. Getting the cover back on the cage was the hardest part, but only took a few minutes. Didn't have the hook tool which would have made the process easier. Once I got the cover back on, everything else was easy and the 6500 is now working fine.
The only tip I have is to keep the old fan and parts well separated from the new ones. They are almost identical. The wires on the new ones are longer than the wires on the originals.
My 6500 was out of warranty and the price of fans was less that the cost of shipping the radio one way.
73 de K1ESE
We are just about at the minimum requirement. At this point I’ll place the order and keep the remaining balance on hand for future need.
I will be placing the order with the factory sometime this week. Lead time is appropriately 7-10 days.
They will be $20 each, plus $10 shipping stateside.
For those outside the US, email me with your shipping addresss for an exact quote. A rough estimate, based on the last time around, is approximately $15 for first class mail, and $25-30 for priority mail.
PayPal to email@example.com
I will reply to everyone ‘s emails later this evening when I get in from work.
I replied to everyone direct via email this evening. Hopefully I haven't missed anyone.
While still somewhat short of the 100 piece minimum requirement at this point, in the interest of time, I have gone ahead and placed the order with the manufacturer. I expect the fans to arrive sometime next week. I will then turn them around within the follow day or two, depending on my work schedule.
If you have not inquired yet, but would like to be on the list, email me at firstname.lastname@example.org