My Fanmate clone

energy · Post by **energy** » Wed Jun 18, 2003 4:13 pm

I've decided the asking price for the Fanmates is too high to actually sell any meaningful amount (I am a reseller), so I've looked into making an identical copy myself.

Prices (for each Fanmate to be built, based on order of 100 quantity):

PCB (exact fanmate clone in green epoxy): 0.8 Euro
Regulator and trimmer (bought both simultaneously so individual prices unknown): 0.95 Euro
0.125W resistor: 0.015 Euro
Large heatshrink: 0.13 Euro per 6cm length
Wires and 2 pin plugs: 0.3 Euro
Misc (screws for heatsink, heatsink compound, solder,glue for trimmer tool): 0.2 Euro

Although I could get hold of similarly sized cases, drilling out all 100 would be too time consuming, apart from having to stabilise the entire PCB in there somehow to stop it moving about. I opted for heatshrink instead - and about 1/10th of the price of a small ABS box.

Unfortunately I ordered 100pc of Bourns trimmers, then later realised they make one with a small shaft, which was what I was looking for. I was intending to get these shafts separately but I cant seem to find them. Might have to toss out the ones I bought and order these instead. Sucks big time as they aren't all that cheap.

Also having trouble finding a heatsink at the same 10x30mm size - anything larger will be a pain to install (heatshrink etc), but it will require some sort of heatsink as the 7805 does get pretty hot even, especially so with limited ventilation (again, heatshrink problem).

Maplin does seem to sell something identical (order number RN87U) but I will have to check with them first (I'm estimating it is the same as the original sink from an overcompressed diagram on their site), and they cost around 60p each which is a damn rip off. I've looked high and low, even Aavid Thermalloy doesn't make them, nor does Rapid, Farnell or RS. Aavid could probably do to order though, and I'm not sure what minimum run they'd want.

Total cost inc Maplin heatsink: £2.30 approx. Assembly will be by yours truly.

Thats about a third of the trade price that QuietPC.com in the UK wants for a Fanmate. Can sell for £5 rather than the usual £10 I have the genuine items for. I will still stock one or two of the geniune ones in case someone wants them but I doubt it

Any ideas where in the UK I can get a reasonably priced (pref. <30p) heatsink like the original from? Anything measuring 10x30mm is fine. Ideally threaded not to mess with nuts and bolts. Also looking for that little trimmer tool to poke out of the heatshrink.

If anyone wants the circuit diagram or PCB trace diagram let me know and I'll post 'em.

Thanks

Zhentar · Post by **Zhentar** » Wed Jun 18, 2003 4:35 pm

A hacksaw and U-channel aluminum should do the trick. you won't be dealing with very much heat on any SPCR type fan.

energy · Post by **energy** » Thu Jun 19, 2003 1:21 am

I'm going to be selling them, so I've no idea what kind of fan someone will be hooking up to them... but yes, might have to use a simple part cut in pieces.

jafb2000 · Post by **jafb2000** » Sat Jun 21, 2003 11:44 am

The U-channel or box section, is probably the easiest to do.

o Re wattage, big 120x38mm fan top out at 7W
o 100cfm 120mm fans top out at 5W

One can assume at most the unit will dissipate 90% of the
maximum fan fitted, a figure of 4W seems reasonable.

o Frankly 4W is nothing to dissipate
---- SMPS metal oxide resistors & coils dissipate that without heatsink
o However touch comfort is a consideration in this application
---- so a limit of 67oC is perhaps best viewed
---- that is the maximum temp generally perceived without pain

That downsizes the heatsink requirement.
A voltage regulator avoids the (many) problems with PWM solutions,
namely frequency used & compatability with the many sizes of fans.
PWM relies on fan slippage being far enough below PWM frequency,
amongst many things but can rapidly create acoustic artefacts at low
fan speed. A voltage regulator by comparison simply varies voltage.

As long as voltage is above minimum for fan start, fine.
--
Dorothy Bradbury
www.stores.ebay.co.uk/panaflofan (Ebay)
http://homepage.ntlworld.com/dorothy.br ... anaflo.htm (Direct Prices)

ravton · Post by **ravton** » Mon Jun 23, 2003 8:42 pm

jafb2000 wrote:o Re wattage, big 120x38mm fan top out at 7W
o 100cfm 120mm fans top out at 5W

Not true. Not even close to true. Maybe the fans you sell fit that but a basic search will find that fans vary much more widely. As an extreme example, I have a 36W 120mm fan in front of me right now.

Jan Kivar · Post by **Jan Kivar** » Mon Jun 23, 2003 11:00 pm

ravton wrote:
jafb2000 wrote:o Re wattage, big 120x38mm fan top out at 7W
o 100cfm 120mm fans top out at 5W
Not true. Not even close to true. Maybe the fans you sell fit that but a basic search will find that fans vary much more widely. As an extreme example, I have a 36W 120mm fan in front of me right now.

What voltage fan is it?

Jan

Kostik · Post by **Kostik** » Tue Jun 24, 2003 12:23 am

There are some 24w and 36w 120mm delta fans, but honestly, who in this forum is going to use a 59db/190cfm rated fan that produces more heat than their cpu ?

Most fan controlers are rated for 7 to 11w. Even if we don't limit our scope to the members of this forum, who would buy a fan > 7w and *undervolt* it ?

energy · Post by **energy** » Tue Jun 24, 2003 2:50 am

As an extreme example, I have a 36W 120mm fan in front of me right now.

As long as it caters to 95% of users fans thats fine. If someone is using a 200CF/M fan and suddenly decides to make it quiet I would be very surprised

jafb2000 · Post by **jafb2000** » Tue Jun 24, 2003 4:15 am

ravton wrote:
jafb2000 wrote:o Re wattage, big 120x38mm fan top out at 7W
o 100cfm 120mm fans top out at 5W
> Not true. Not even close to true. Maybe the fans you sell fit that but a
> basic search will find that fans vary much more widely. As an extreme
> example, I have a 36W 120mm fan in front of me right now.

Ravton, try reading what I write.

o 100cfm 120mm fans top out at 5W
o Big 120x38mm fans top out at 120-128cfm, and around 7W

In the context of what design spec re heatsink/market fit.

YES:
o A FEW 120x38mm fans will go *much* higher
o FEW = less than 1% of 120mm fans available
o MUCH higher = 7W gets 128cfm, but *24W* required for 190cfm
---- that is a /disproportionate/ increase in wattage

You cited a Delta 190cfm fan re 36W at 120mm.
o that is incorrect - it is a 2.0A continuous (24W) fan
o only startup is 3.0A (36W), which is not used for HVAC calculations

Essentially one has to choose a design limit - I cited 100cfm & 5W.
o Sure, you can go to 120-128cfm and get to 7W
---- however that is a <50% increase on 5W
o However much above 100cfm leaves you few fans of *>>>5W* (24W)
---- 24W is a design requirement some 500% higher than 5W

Design must encompass the most buyers economically:
o How many buyers will buy >>5W 100cfm fans?
---- to pay the higher premium for such fans
---- only to then drop them down to 50% of that airflow re noise
o How can one justify a voltage solution for >>5W fans?
---- this isn't a Mosfet PWM solution
---- this is a voltage regulator + heat + heatsink solution
---- heat which the fan must them remove from the case
---- for a 24W fan this is potentially 12W more in the case
o How much extra does such control capability cost?
---- the jump is from a 5W limit to 24W continuous, 36W startup
---- that's from 2.5W dissipated to 12.5W dissipated
---- for an ambient of 25oC, surface of 69oC, that's ~3.5oC/Watt
o Fitting a 3.5oC/Watt heatsink vs 22oC/W for the top 5% of buyers
---- so 95% of buyers must pay more for a bigger heatsink & size
---- or your margins just got eroded to catch the last 5% of buyers
---- a 3.5oC/Watt heatsink is physically much bigger than 22oC/W

There is arguably some internal case airflow to assist the heatsink.
However, the item may be placed in "dead-space" and get <2cfm.

There is then the control achievable over such a 24W fan:
o The spec is 57dB(A) in free air, so 62-67dB(A) fitted
o Running at 5V the best one is likely to get is ~30dB(A)
o Running at 7-8V and we're nudging 37-39dB(A)

So the controller can only achieve "quiet" for such a fan.

So the argument is do you support 5W or 7W.
I don't see an argument for supporting 24W continuous fans,
because a voltage regulator isn't suitable for that solution.
Considering a 235cfm 37W fan exists (OEM) it gets even sillier.

Supporting beyond 5W or 7W pushes the product into PWM solutions:
o From the PC retail market
o From PAPST re PCM001 & System3000 (25W & 65W)
o From Control Resources (15W to 200W if I recall, www.controlres.com)
o From the more HVAC orientated market

Design must consider a limit, which is much nearer 5W than 24/37W.
Marketing must consider a limit, re which theatre to compete in.

Big 120x38mm fans top out at 7W for 99% of user-base.
The 100cfm fans top out at 5W, and frankly is a reasonable goal.

Supporting beyond 7W is self-defeating:
o The controller can't make them particularly silent
o The controller has to have a heatsink considerably larger

Frankly the same applies to beyond 5W for compact sizing & cost,
at least where the assumption is of no airflow cooling over the device.

ravton · Post by **ravton** » Tue Jun 24, 2003 6:13 am

jafb2000 wrote:Ravton, try reading what I write.
o 100cfm 120mm fans top out at 5W
o Big 120x38mm fans top out at 120-128cfm, and around 7W
In the context of what design spec re heatsink/market fit.

I read what you wrote exactly, apparently you did not. You did not specify any contect, you may have assumed it, but you certainly did not specify it.

jafb2000 wrote:YES:
o A FEW 120x38mm fans will go *much* higher
o FEW = less than 1% of 120mm fans available
o MUCH higher = 7W gets 128cfm, but *24W* required for 190cfm
---- that is a /disproportionate/ increase in wattage

Spare us your created-on-the-spot statistics please.

jafb2000 wrote:You cited a Delta 190cfm fan re 36W at 120mm.
o that is incorrect - it is a 2.0A continuous (24W) fan
o only startup is 3.0A (36W), which is not used for HVAC calculations

Incorrect. A current Delta is 2.0A continuous, an older model is spec'd at 3.0A continuous. Actual tested draw is about 2.6A at about 12V. As I said, this was cited as the extreme example, to illustrate that your statements are not factual.

jafb2000 wrote: Essentially one has to choose a design limit - I cited 100cfm & 5W.

Yeah, but you didn't say that. You merely categorically said "Big 120x38mm fans top out at 120-128cfm, and around 7W". Clearly not a truthful statement.

Being that you're a DEALER of fans, you should be more careful of what you claim to be facts.

ravton · Post by **ravton** » Tue Jun 24, 2003 6:15 am

energy wrote:
As an extreme example, I have a 36W 120mm fan in front of me right now.
As long as it caters to 95% of users fans thats fine. If someone is using a 200CF/M fan and suddenly decides to make it quiet I would be very surprised

I didn't say that you should. As stated, I merely used this an example to illustrate that the facts that the fan dealer stated are not true. I certainly agree that using a linear fan controller with a 36W fan is not the best move

.

jafb2000 · Post by **jafb2000** » Tue Jun 24, 2003 6:23 am

> Incorrect. A current Delta is 2.0A continuous

Which is what I am referring to.

Regarding statistics:
o The original post was regarding a Fanmate competitor
o The Fanmate has a limit of 7W (altho somewhat warm)
o Thus my reply was correct in context as I stated

Other posters also replied similarly regarding that context.

The conversation is thus somewhat pointless:
o You have made a comment
o I have replied to it

I stand by my reply. You stand by yours.
--
DB.

ravton · Post by **ravton** » Tue Jun 24, 2003 6:52 am

jafb2000 wrote:> Incorrect. A current Delta is 2.0A continuous
Which is what I am referring to.

Regarding statistics:
o The original post was regarding a Fanmate competitor
o The Fanmate has a limit of 7W (altho somewhat warm)
o Thus my reply was correct in context as I stated

Other posters also replied similarly regarding that context.

The conversation is thus somewhat pointless:
o You have made a comment
o I have replied to it

I stand by my reply. You stand by yours.
--
DB.

*sigh*

I never should've mentioned the 2.0A fan as you latched onto that to try to disguise your unfactual statements.

Again, you stated facts that are simply not true nor even close to true.

You said:

o Re wattage, big 120x38mm fan top out at 7W
o 100cfm 120mm fans top out at 5W

This is categorically untrue. I don't see why you are defending this.

jafb2000 · Post by **jafb2000** » Tue Jun 24, 2003 10:29 am

I'll repeat it again:

o The threads title is "My fanmate clone"
---- my post is with regard to the design of such
---- specifically what wattage assumption to make
---- and thus what sizing of heatsink to meet it

o I stated:
---- "o Re wattage, big 120x38mm fan top out at 7W"
---- "o 100cfm 120mm fans top out at 5W"

For purposes of designing "My fanmate clone":
o Fanmate chose a 7W limit for their device
o The big 120x38mm fans top out at 7W also
o 100cfm 120mm fans top out at 5W

Thus for heatsink sizing 5-7W is logical.

YES, *some* 120x38mm fans top out at 24-36W.

Sizing for the few 24-36W fans is illogical:
o Much larger heatsink (~2.2oC/W v 22oC/W)
o More expensive product or lower margins
o There are few 24-36W fan buyers
o There are even fewer 24-36W fan buyers who then after paying up for
such a fan THEN pay up to reduce the fan back to cheaper fan cfm levels
o There are fewer who want 12W extra heating from using a voltage
regulator for a 24-36W fan control

For design purposes 7W, if not 5W, is a good limit.
The number of fans exceeding 5W is low, 7W even lower.
To incorporate fans above 5W has low credibility.

Which hilariously shows how incredible your comments at taking
mine out of context are.

Sorry if too much was not made explicit, but THAT is why I replied
as I did - for very good reason.

So I'll spell it out:
o For the purposes of "My Fanmate Clone" topic
o My statement is AS I INTENDED re heatsink sizing
o Big 120x38mm fans top out at 7W
---- for INTENDED FANMATE CLONE BUYERS
o Even 100cfm 120mm fans top out at 5W
---- which is true

Both statements are correct in their context.

Both statements are incorrect out of context.
o Big 120x38mm fans top out at over 7W
---- they actually go beyond 36W to 57W
o 100cfm 120mm fans top out at over 5W
---- because 114mm deep fans exist at 72W

I made the statements I did for very good reason,
and those reasons are made very explicit above.

You can argue the statements are untrue until hell freezes over and
Saddam then does walk on water. Out of context they are not correct,
in context of fan user-population for a fanmate clone design they are.

It's a stunningly pointless argument. Hilarious though.

ravton · Post by **ravton** » Tue Jun 24, 2003 6:00 pm

You're the only person arguing that a 24W design is unreasonable; I hope you're having a good time.

The facts again are just that you claimed that 120mmx38mm fans top out at 7W. Something that is clearly not true. My post corrected what you said.

You created and have continued the comedy by continuously restating that your misspoken remarks were intentional. Funny this is, but I don't think you've figured out yet why it's funny.

Post by **Rusty075** » Tue Jun 24, 2003 6:31 pm

Whoa, Ravton, back it down.

I think anyone who reads this thread, here on Silent PC Review, would assume that Energy was talking about building a device to clone a Fanmate. The device clearly was not met to be used with any fan ever made. (That's why Fanmates have a little disclaimer on them)

Therefore Jafb2000 was being perfectly logical in basing his technical input on the assumption that a "normal" fan was what the device was being designed for.

Your whole argument could have been phrased like this, "Some fans have much higher wattage draws than that, perhaps the device should be made to accomodate them" At which point Energy, or Jafb2000 could have responded with their theories as to why or why not that should be.

And then a useful discussion could have started as to whether or not this "product" of Energy's should be aimed at PC silencers only, or the rest of the computing market as well.

But you didn't.

It's a non-issue. Walk away. Do not let your self-asteem is determined by posts in a forum.

ravton · Post by **ravton** » Tue Jun 24, 2003 6:54 pm

Rusty075 wrote:Whoa, Ravton, back it down.

Woah. Where'd that come from? I don't think I've gone anywhere to back down from.

Rusty075 wrote:I think anyone who reads this thread, here on Silent PC Review, would assume that Energy was talking about building a device to clone a Fanmate. The device clearly was not met to be used with any fan ever made. (That's why Fanmates have a little disclaimer on them)
Therefore Jafb2000 was being perfectly logical in basing his technical input on the assumption that a "normal" fan was what the device was being designed for.

But that isn't what he/she said. He/she make a categorical statement that 120mm x 38mm fans MAX out at 7W which is not true. I tried to help out by stating that this "expert" is incorrect. Everything else that has been said by he/she is based upon arguments that I never said nor even insinuated. Rather bizarre is what I've been thinking.

Rusty075 wrote:Your whole argument could have been phrased like this, "Some fans have much higher wattage draws than that, perhaps the device should be made to accomodatethem" At which point Energy, or Jafb2000 could have responded with their theories as to why or why not that should be.

That was never my point. Read my post again, I said "Not true. Not even close to true. Maybe the fans you sell fit that but a basic search will find that fans vary much more widely. As an extreme example, I have a 36W 120mm fan in front of me right now."

Rusty075 wrote:It's a non-issue. Walk away. Do not let your self-asteem is determined by posts in a forum.

Heh, Nothing could be farther from the truth. Read through the posts again and figure out who has gotten emotional here.

You know, I didn't intentionally try to bait him/her, Him/her took this upon themselves. I really liked the long diatribe though.

LushMD · Post by **LushMD** » Tue Jun 24, 2003 7:33 pm

ravton: considering that I am far from being an expert re such matters, perhaps I shouldn't comment. However, as an SPCR member that has much respect for the knowledge and opinions of both Dorothy (jafb200) and Rusty, I must agree that your tone is rather inflammatory. To wit, your contributions to this thread are not properly phrased to engender collegial conversation/discussion, which is the crux of any meaningful forum. Take care.

cpemma · Post by **cpemma** » Mon Jul 14, 2003 2:01 pm

jafb2000 wrote:...As long as voltage is above minimum for fan start, fine.

I gather from someone who bought and opened up a FanMate that the circuit is a 7805, a fixed resistor and a pot, so (a) at over £8 it's a rip-off, and (b) it can't go below 5v, which is a bit borderline but should start most fans.

On heat-sinks, worst case is a 5.125v drop at 0.5A (to allow the 10.25v upper limit at 1A), 2.56W, so for say 50C rise in temperature, 50/2.64 less the 4.3C/W for the TO220 package gives a 15C/W sink. I don't know what the RN87U spec is, but Maplin's RN77J at the same price is 13.5C/W, and I'm sure RS Components will have something similar and cheaper (but that's a meaty HS and would run 14W of fannage).

Setting a sensible recommended limit on fan wattage, you can ease up a lot. ESR do 17C/W TO220 sinks at 17p (100+).

Phill77 · Post by **Phill77** » Mon Jul 14, 2003 2:30 pm

Cpemma, surely you have just missed a golden opportunity to plug the wonderful little 555 PWM circuit you have on your website.
Always hits the fan with 12V (ish), so no problems starting and it doesn't seem to throw much heat out.

Cost me four quid including a nice knob and took an hour of my time. Down with fanmates!

energy · Post by **energy** » Mon Jul 14, 2003 4:11 pm

- Turns out that 0.125W resistor shaped thing is not a resistor after all - its a smoothing capacitor on the output of the regulator. The system works fine without it.

- I since dropped the idea of making a clone - finding identical size parts and encapsulating them was too difficult, not to mention having the wrong trimmers (lack of knob/shaft).

- Instead I decided to make it using stripboard, any old small heatsink and encapsulate it with potting compound, leaving the trimmer sticking out. Total cost: £1.50 and very small. Assy takes 3 minutes. Only fiddly thing is the 3 pin connector crimps, thay are a pain in the behind.

- Luckily quietpc.com has cut a chunk off their trade prices and fanmates are approx £1.00 cheaper now, which is something. Other products are also cheaper.