VOGONS


First post, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie

I have two of these,

https://www.digikey.com/en/products/detail/de … 1212UHE/2560583

They were basically the holy grail of fans when my buddies and I were messing with computers, my friend even bought one and fed it pencils ✏️ 🤣 never broke a blade if you can believe it haha.

Anyway, I think they are cool as shit and want to install them in my build. But problem is, no way in hell do I want to feed them full voltage haha.

I also have one of these (soon)

https://www.ebay.com/itm/8-Channels-PC-Case-C … be-ede6c6901e54

It looks like it has four fets with two fan ports each.

Could I install my cpu fan on the first fet channel, have the first delta on the second fet channel, the second delta on the third fet channel, and leave the remaining fet channel empty?

Do those fets look beefy enough to take those fans?

What if I feed the controller only 5v?

My other option is to simply not plug them in but I hate that 🤣. If I can just tone them down that would be ideal. It’s not like I need them. I don’t even need this build. It’s all for fun 🤣 I’d like them to at least spin slow.

Attachments

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 1 of 19, by computerguy08

User metadata
Rank Member
Rank
Member

I'm guessing putting 48W Delta fans inside a small computer has no practical reason, other than looking badass.

Connecting just one of them to that controller will make it let out the magic smoke. Even at 5V, I don't think it can handle them.

Your best option is to hot wire them on 5V directly or grab some less hungry Delta fans.

Looking for a motherboard? You can find it in Ultimate Hardware 2019: http://www.win3x.org/uh19/motherboard/search
Join our UH19 Discord server here: https://discord.gg/HWWH7hsk2p

Reply 2 of 19, by adalbert

User metadata
Rank Member
Rank
Member

Maybe you could use external PWM controller like this one https://www.ebay.com/itm/PWM-DC-5V-16V-10A-DC … l8AAOSw2uhey3GO

it just would need to be wired manually. The controller you linked earlier probably uses linear voltage regulators and creating lots of waste heat, this one will waste much less power. It will just create longer or shorter pulses of 12V.

Reply 3 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
computerguy08 wrote on 2021-01-27, 14:03:

I'm guessing putting 48W Delta fans inside a small computer has no practical reason, other than looking badass.

Connecting just one of them to that controller will make it let out the magic smoke. Even at 5V, I don't think it can handle them.

Your best option is to hot wire them on 5V directly or grab some less hungry Delta fans.

Maybe if I put the fans in series.

Or I could make a combiner, combine three channels into two ports and the ports themselves are in series. 🤔

Idk, I guess I just need to try it. Plug it in real quick, see if they get warm at low setting. Start with 5v

Should be able to figure out if they are too much by how warm the fets get I’d think.

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 4 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
adalbert wrote on 2021-01-27, 14:28:

Maybe you could use external PWM controller like this one https://www.ebay.com/itm/PWM-DC-5V-16V-10A-DC … l8AAOSw2uhey3GO

it just would need to be wired manually. The controller you linked earlier probably uses linear voltage regulators and creating lots of waste heat, this one will waste much less power. It will just create longer or shorter pulses of 12V.

Good idea, I can make my own bracket with those. Won’t look as nice though.

So those regulators are just adding resistance? Where as the pwm turns it on and off fast?

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 5 of 19, by adalbert

User metadata
Rank Member
Rank
Member
Sphere478 wrote on 2021-01-27, 15:00:

So those regulators are just adding resistance? Where as the pwm turns it on and off fast?

Probably thats correct, because I don't see any kind of 'smart' chip on the controller you linked. I can't read the markings on these four chips though so I don't know for sure.

PWM controller switches very fast, usually 1kHz or faster.

Reply 6 of 19, by frudi

User metadata
Rank Member
Rank
Member

Those fans have an operating voltage range of 8 - 13.2 V, according to the spec sheet. They probably won't even start up at 5 V, perhaps at 7 V. They will be unbearably loud even at that speed, take it from someone who used to run an 80 mm Vantec Tornado at 7 V back in the day. Unless you have genuine hearing damage, you probably don't want to run these, even at minimum speed.

Reply 7 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
frudi wrote on 2021-01-27, 21:05:

Those fans have an operating voltage range of 8 - 13.2 V, according to the spec sheet. They probably won't even start up at 5 V, perhaps at 7 V. They will be unbearably loud even at that speed, take it from someone who used to run an 80 mm Vantec Tornado at 7 V back in the day. Unless you have genuine hearing damage, you probably don't want to run these, even at minimum speed.

you're probably correct I fear. 🙁 still gonna try and see for sure.

ah I forgot about the tornados ah the good ole days

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 9 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
adalbert wrote on 2021-01-28, 01:39:

PWM control should allow fan startup at lower speeds than usual, because in theory you are not reducing the voltage, but you are reducing the duty cycle.

Recommendation of a bracket like this one that has that?

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 10 of 19, by Doornkaat

User metadata
Rank Oldbie
Rank
Oldbie

PSU manufacturers don't want you to know this but there's actually a 7V rail on your standard 4pin 5.25"-drive connectors.
Measure for yourself, there's a ~7V difference between the outer wires.
Just keep in mind that even at 7V the fan is still drawing ~2.33A/16.33W. 😉

Reply 11 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
Doornkaat wrote on 2021-01-28, 07:11:

PSU manufacturers don't want you to know this but there's actually a 7V rail on your standard 4pin 5.25"-drive connectors.
Measure for yourself, there's a ~7V difference between the outer wires.
Just keep in mind that even at 7V the fan is still drawing ~2.33A/16.33W. 😉

It’s a clever trick. Indeed, there are in fact many
voltages possible between 3.3, 5, -5, 12 -12

You can go between gnd and -12 and get -12
You can go between -12 and -5 and get -7
You can go between gnd and -5 and get -5
You can go between gnd and 3.3 and get 3.3
You can go between gnd and 5 and get 5
You can go between 3.3 and -5 and get 8.3
You can go between 3.3 and 12 and get 8.7
You can go between -5 and 5 and get 10
You can go between gnd and 12 and get 12
You can go between -5 and 12 and get 17
You can go between -12 and 12 and get 24
Including more negative options By reversing the leads

Is there a -3.3 now also? I forget? Anyway. You can see how the atx power supply is handy for making complicated circuits like computers haha

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 12 of 19, by wiretap

User metadata
Rank Oldbie
Rank
Oldbie

There's always the option of using 120mm muffin fans that run on 120VAC power. I use those in quite a few systems at work. Some of them have metal serrated blades which could chop off a finger. We have rheostats (like a house lighting dimmer switch) hooked up to them to control fan speed. Noise isn't really a factor since they are either used in a server room or industrial plant setting.. but they defintely scream and move large amounts of air.

Circuit Board Repair Manuals
My Project List

Reply 13 of 19, by Doornkaat

User metadata
Rank Oldbie
Rank
Oldbie

Even though running a fan at 7V by hooking it up between the two positive rails on the PSU is an old trick and you can (or could?) actually buy commercially produced adaptors that work exactly like this I do not know wether this can cause any problems on some PSU designs because it is not how they were intended to run.
If you really want to run a fan like those Deltas in your system I would recommend using either a PWM controller like already sugested or a (variable?) resistor, though I doubt you'll find a ready solution or either designed to work in a PC.
Here's a link that'll help you in calculating the resistor you need if you choose that option.

Reply 15 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
wiretap wrote on 2021-01-28, 15:16:

There's always the option of using 120mm muffin fans that run on 120VAC power. I use those in quite a few systems at work. Some of them have metal serrated blades which could chop off a finger. We have rheostats (like a house lighting dimmer switch) hooked up to them to control fan speed. Noise isn't really a factor since they are either used in a server room or industrial plant setting.. but they defintely scream and move large amounts of air.

I actually have one or two of those. The point was these fans though for their look and personal history. It may come down to just not plugging them in as they really are not needed

Hoping wrote on 2021-01-28, 19:11:

I think you'll find this page interesting http://www.pcsilencioso.com/cpemma/ef.html ,there's some info on making fan controllers.

Oh very nice! 😀 thanks!

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 16 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie

okay, I got that controller it will fire up the fans but the fets start to get hot when you turn up the speed also the fans stall when you turn down the speed I'm convinced if I leave it on in any setting that something will fry the chips really did start getting hot and the fan stalling is obviously a no go these fans don't even seem to like straight 5v I'm convinced now after testing that the way they are designed (I think there is onboard circuitry) is such that anything less than 12v (or close there to) will cause the fans to operate abnormally and possibly damage them so anyone reading this just give up, the controller can't take them and the fans don't like being turned down. sorry,

I'm gonna leave em unplugged or get different fans. the controller does seem to work well on other fans and seems to be decent quality. this is probably the case on the 220cfm model as well

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 17 of 19, by pentiumspeed

User metadata
Rank Oldbie
Rank
Oldbie

UPDATE

Your fans is not the problem. Linear regulators works well with 3 pin fans. Back in the day, I created well designed regulator on a experimental PCB with many components for 3 pin fans to quieten down computer.

I investigated the fan controller you linked recently as I'm still looking for *true* manually adjustable PWM signal controller on a bracket and turns out there is no extra components required to generate PWM signal. These "fets" are actually linear voltage regulators dialed by the potimeter. This *does not* work with PWM fans especially greater than 1A fans as they sped up to full speed no matter what.

Currently Noctua only have this one but it is expensive at 40 to 50 a pop. Chinese trinkets on the market does have real PWM adjustable but this is bare board, not even in a box, temperature (I have two and I used it without sensors to evaluate my collection of PWM fans for specific purpose (air pump with pressure and hose, to inflate plastic bags) and push buttons (complicated way to adjust) not knob not even assembled to a bracket. Disadvantage is it's microcontroller, disadvantage two, no selectable frequency like 25KHz or 16KHz, third disadvantage is no switch to invert a PWM signal.

People are making 1-100% PWM controllers set by a pot using two 555 timers and LM113, with kick start feature added on, negative thing is nobody had designed boards singles and multiples to sell bare boards so anyone can solder the parts on. Also designed to use project boxes. This also allows you to change frequency as well by swapping a capacitor and add a simple circuit to invert PWM signal.

The attributes of server fans are very particular about PWM signal requires this.

I would do that too as I plan to make one. They are on overlockers forums. I'll link these direct to there after work, I'm on work computer which does not have the links.

As promised, here's the link to schematic to make the true PWM signal output and it is heavily tested and researched as bing needed to run server fans properly. the one 1-100% works best, and few notes on how to add features and change either 25KHz or 16KHz is one capacitor change.

https://www.overclockers.com/forums/showthrea … 318#post6473318

Cheers,

Great Northern aka Canada.

Reply 18 of 19, by Sphere478

User metadata
Rank Oldbie
Rank
Oldbie
pentiumspeed wrote on 2021-02-02, 15:36:
UPDATE […]
Show full quote

UPDATE

Your fans is not the problem. Linear regulators works well with 3 pin fans. Back in the day, I created well designed regulator on a experimental PCB with many components for 3 pin fans to quieten down computer.

I investigated the fan controller you linked recently as I'm still looking for *true* manually adjustable PWM signal controller on a bracket and turns out there is no extra components required to generate PWM signal. These "fets" are actually linear voltage regulators dialed by the potimeter. This *does not* work with PWM fans especially greater than 1A fans as they sped up to full speed no matter what.

Currently Noctua only have this one but it is expensive at 40 to 50 a pop. Chinese trinkets on the market does have real PWM adjustable but this is bare board, not even in a box, temperature (I have two and I used it without sensors to evaluate my collection of PWM fans for specific purpose (air pump with pressure and hose, to inflate plastic bags) and push buttons (complicated way to adjust) not knob not even assembled to a bracket. Disadvantage is it's microcontroller, disadvantage two, no selectable frequency like 25KHz or 16KHz, third disadvantage is no switch to invert a PWM signal.

People are making 1-100% PWM controllers set by a pot using two 555 timers and LM113, with kick start feature added on, negative thing is nobody had designed boards singles and multiples to sell bare boards so anyone can solder the parts on. Also designed to use project boxes. This also allows you to change frequency as well by swapping a capacitor and add a simple circuit to invert PWM signal.

The attributes of server fans are very particular about PWM signal requires this.

I would do that too as I plan to make one. They are on overlockers forums. I'll link these direct to there after work, I'm on work computer which does not have the links.

As promised, here's the link to schematic to make the true PWM signal output and it is heavily tested and researched as bing needed to run server fans properly. the one 1-100% works best, and few notes on how to add features and change either 25KHz or 16KHz is one capacitor change.

https://www.overclockers.com/forums/showthrea … 318#post6473318

Cheers,

You know what would be cool, is if you or someone could make a pcb that replaces the one I have, I could unsolder the POTs and the headers and transplant them to the new board. Any chance of doing that?

🖥Craziest socket 7 build on a 430tx chipset
🖥Dual socket 7 build

Reply 19 of 19, by pentiumspeed

User metadata
Rank Oldbie
Rank
Oldbie

Problem is like I stated, I have not yet to see a PCB designed and available for sale to solder parts on to make PWM board. No, I don't want to re create 6 times circuits by hand on experimenter's board to drive different speeds for specific PWM fans.

Once I validate the circuit, I want to have a PCB made to look nice with all the features included.

Cheers,

Great Northern aka Canada.