I am having am issue with a Swiftech 8 way PWM block

beekermartin · Sep 20, 2016

I installed the block last night. Sata cable is for power and the 2 wire 4 pin connector is connected to the CPU PWM 4 pin connector on my motherboard. It is working but only the fan on channel one is being controlled by the PWM signal. The other 4 fans are all spinning at max speed. I tried moving the fans around and they all work properly if connected to the #1 channel or directly to the mother board. If they are connected to any of the seven other channels they spin at full speed. I had read that this block has issues working with Corsair fans but I am not using Corsair fans. I am using Cougar fans.

Is the block I have faulty? Should I return it and try another one? Swiftech tech support wasn't much help. They didn't seem to know if the issue is the block or it not being compatiable with my fans. They were willing to RMA but I will probably just return it to Amazon because it will be easier and faster.

Here is a link to the block: http://www.swiftech.com/8-WayPWMsplitter-sata.aspx

bigpinkdragon286 · Sep 20, 2016

I doubt that Corsair is doing anything special with their fans that make them incompatible. A PWM controlled fan is just that, so it seems unlikely to be a branding issue. If the fan controller is problematic with Corsair, it's likely the same for all.

Here's a question though, if you're only connecting to a single PWM header on the motherboard, are you expecting all of the headers on the fan block to run with the same PWM signal from the one fan header, or are they actually supposed to run at full speed, except for the one that is working?

Edit: the Swiftech link explains that all headers are supposed to run at the same speed, but only #1 carries the tach signal for the motherboard fan header.

Are all of the fans the same make and model? If they are designed to run at different RPMs for a given PWM signal, that may be affecting your speeds.

Also, the Swiftech site mentions that the PWM block is only compatible with a PWM implementation that has a constant +12 V and a separate PWM signal. Some PWM controllers apparently modulate the +12 V to vary the fan speed, which isn't the correct way to run a PWM fan.

If there is an option in your motherboard's BIOS to use voltage control instead of PWM control for your CPU fan, make sure it is set to use PWM control.

beekermartin · Sep 20, 2016

They are all suppose to run at the same speed from the one header on the motherboard. That is the idea of the block. To be able to connect up to 8 identical fans and have them all run at the same speed based on the PWM signal of the motherboard header it is connected to.

The issue, I read, with the Corsair fans had something to do with them requiring a higher than normal PWM signal and that caused them to not work with the Swiftech block. I have no idea if that is true but that is what I had read.

beekermartin · Sep 20, 2016

I will check the bios settings of the CPU fan header and make sure it is set to PWM. Keep in mind that the block is working for the single fan connected to channel 1 of the block. Which means the block is working but not duplicating the signal to the other 7 fans.

In my opinion the block is defective. I was hoping someone else here might be using this block with Cougar fans. That way I know if it is a compatibility issue like with Corsair fans or if the block is in fact defective.

I have two Coolermaster Blademaster fans. I am going to try them connected to it to see what happens. If only one is controlled like with the Cougar fans than I will know the block is defective.

bigpinkdragon286 · Sep 20, 2016

Upon further reading, your profile says you're running a Gigabyte Z97 motherboard. This is likely your issue. There is a great Amazon review that states:

"Works perfect. As long as you dont have a Gigabyte Z87/97 Mobo. If you do have a z97 or 87 the only channel the board will allow you to control by PWM is the CPU fan header. Any other header this is hooked to on a z97.87 mobo the fan will run at full speed on the splitter. The reason is the gigabyte boards for some really stupid reason require the fan header to sense 12 volt before it will allow PWM control. And since splitters only use the last two wires on the header the board never senses 12 volts and will not output a PWM control signal.
Have always liked Gigabytes and have had good luck with them, but this will be my last for this reason."

May not be the block or fans at all. It may be the way the motherboard is communicating to the block. If the motherboard is not sending it a proper PWM signal, the block will likely malfunction in some fashion.

Are you connected to the CPU fan header or one of the auxiliary fan headers?

beekermartin · Sep 20, 2016

bigpinkdragon286 :

I am connected to the CPU header. Also, why would PWM work for a single fan and not the rest if the above quote is accurate? If the issue was the motherboard I don't believe PWM would work for any of the fans but I could be wrong. I will know more once I test a few other fans. I will post again after I've done that.

bigpinkdragon286 · Sep 20, 2016

The very next review on Amazon explains the Corsair issue:

"It does split the PWM signal as expected, but does nothing to amplify it. When connecting my Corsair SP120 High Performance editions I am only able to connect two of them before the signal gets too weak, causing *all* connected fans to run full speed. Supposedly this is due to SP120 needing strong PWM signal, but it does turn this into essentially a 2-way splitter with these fans. Your mileage may vary."

Basically, the Swiftech is just splitting the PWM signal without any method to boost it. There is nothing wrong with the Corsair fans, they simply don't see the signal after it's been split too much, which is not something unusual. No different than splitting something like a TV or radio signal. There's only so much signal available, so if you split it far enough that it falls below your minimum threshold, you're out of luck.

The splitter seems to work with 2 Corsair fans attached, so it's not completely worthless in that regard. Also, if signal splitting was the issue, you could probably run just a single of your fans on a different header than #1 and it should receive enough PWM signal to function correctly, you just wouldn't get a tach signal for the test. You would probably have to turn off CPU fan failure protection in BIOS to try this though.

bigpinkdragon286 · Sep 20, 2016

It may be that the CPU header is not actually outputting a PWM control signal, but attempting to perform voltage control instead, and the Swiftech block may be passing that through to the #1 fan header somehow, but not the rest.

beekermartin · Sep 20, 2016

Thanks for all the responses! I will double check the CPU header setting in the bios when I get home. I have been using a normal two way splitter with either the Cougar or Blademaster fans and the PWM function of the CPU header worked properly.

bigpinkdragon286 · Sep 20, 2016

A normal splitter should work for any motherboard header, if it uses all four wires. It sounds to me like Gigabyte automatically switches the four pin fan headers to a different mode when they don't think they see a PWM fan, as a convenience feature I'm sure, but is also where the problem lies with the Swiftech controller. It doesn't appear as a proper PWM fan.

If Gigabyte implemented the auto sensing on the CPU header as well in later BIOS revisions, this would certainly be problematic. I don't see why they would have one PWM header that operated differently than the others, but the Amazon reviewer certainly seemed to think that was the case.

beekermartin · Sep 20, 2016

The Coolermaster Blademaster fans do the exact same thing as the Cougars. The fan connected to channel 1 is controlled by the PWM setting of the CPU header. The remaining 7 channels all run at full speed. I checked the bios and all the fan headers, except the OPT Cpu header, are set to normal. The OPT header is set to full speed because it is currently powering a H60 pump. I will be upgrading to a custom loop soon which is why I need a fan splitter.

There is no option in the bios to change from voltage or PWM. It appears all fan headers are PWM. There are options to set the percentage of PWM/C but I am using software in windows to control the PWM function so normal is the correct selection. Again, I don't believe this issue is the motherboard only having 2 wires connected to the CPU header. If that was the case then none of the channels, including channel 1, would have PWM control. Since channel 1 is working properly that means to me the PWM function of the motherboard is working properly. The block isn't transmitting that signal to the other 7 channels like it is suppose to do. Since I have tried two different brand fans and gotten the exact same results I have eliminated the fans as the issue as well.

I am going to return the block and get a replacement. Hopefully I am correct and the replacement works as it should.

beekermartin · Sep 20, 2016

I just ordered a replacement block. I will post again after I get to test it. Unfortunately that won't be until next Monday at the earliest. I am leaving town Thursday am and won't be back until Monday.

bigpinkdragon286 · Sep 21, 2016

No problem, we'll still be here when you get back. Hopefully the new block sorts things out, but I have my suspicions it may also behave the same as the old block does, and I really suspect the low cost design of the block to be part of the problem.

beekermartin · Sep 21, 2016

You are probably correct. If the next one does the same thing I will have to find something else. Most reviews about this one have been positive. I guess it could be my motherboard but I don't see that being the issue since channel one is being controlled properly. I guess I will find out soon enough. Thanks for your help!

bigpinkdragon286 · Sep 21, 2016

It's not a bad looking unit, and I've even looked at it myself as I'm in the market for a low cost device to drive some higher amperage fans, but ultimately, I think I'm going to have to just bite the bullet and wire my fans to a power attachment manually.

Edit: Found what I was looking for here, basically power comes from the power supply rather than a fan header so high draw fans can be used, pass through for the PWM signal, and a single tach connection.

beekermartin · Sep 26, 2016

I received the replacement Swiftech 8 way PWM block and it is working properly. All channels are running at the PWM setting using windows software or the bios settings. I have them all set to normal in the bios and using Gigabyte System Information Viewer software to control the fans. I might switch to Speedfan but the Gigabyte software does the job.

So the original Swiftech 8 way PWM block I received was defective. The replacement is working properly and as advertised.

bigpinkdragon286 · Sep 27, 2016

Glad to hear you got it working! Thanks for the update as well.

beekermartin · Nov 23, 2016

I am going to post a quick update just in case someone else stumbles upon this thread in the future.

I have spent more time than I care to admit trying to figure out issues with my 4 pin fan headers on my Gigabyte mobo. Long story short, not all 4 pin headers are actually PWM headers. Absorb that for a second....Ok, for some odd reason mobo manufacturers started putting 4 pin headers all over their motherboards. Unfortunately they don't tell you that only a few, or one in my case, is an actually PWM header. The rest are just three pin non PWM headers with an extra useless pin added.

The reason I am posting that here is I am not 100% positive I was using the single PWM header on my mobo when I first tested the Swiftech splitter. I believe I was because I didn't change anything when I received the replacement, which worked perfectly, but I am not 100% sure on that. So I could have been using the CPU Opt header with the original Swiftech splitter and that would explain why it wasn't working. Only the CPU fan header is a true PWM header on my mobo. I believe I did test it on the CPU header before I returned it but again I cannot be 100% on that.

So if you get the Swiftech splitter and it isn't working double check your mobo manual to see if the header you are using it on is an actual PWM header. You may have to call the manufacturer to verify because the bios may even say it is PWM but it might not actually be. I guess PWM never really caught on so manufacturers started to drop it from a lot of headers to save a few pennies. Why they kept the 4 pin header without PWM functionality is beyond my understanding. I wish they wouldn't do this but they do. Gigabyte isn't the only one. A lot of mobo manufacturers are doing the same thing.

bigpinkdragon286 · Nov 23, 2016

This goes back to the findings of the original reviewer I posted, where the author noted for certain Gigabyte motherboard lines, including the one in your sig, the only fan header that is going to give you a PWM signal with that block is the CPU fan header.

I wouldn't write off the other fan headers as non-PWM before testing them the way they are intended to be used, with a single, PWM fan.

Did you ever try running a single PWM fan from the fan headers that would not output PWM signals through your splitter block?

I highly suspect that Gigabyte's engineers know what they're doing, and corporate would not spend the money on the larger fan headers unless there was a need for them.

beekermartin · Nov 24, 2016

I have tested the other headers using not only the Swiftech splitter, but also an EK D5 PWM pump (which has only the third and fourth wires connected, like the Swiftech splitter, because it gets its power from a separate power supply header) and also individual fans. After spending hours of research and testing I am 100% positive that my motherboards cpu header is the only true PWM header. The rest are 4 pin headers but I have no idea why. Maybe some devices require a signal on the fourth pin so they include it. All I do know is they are not PWM.

Do some searches for 4 pin headers that are not true PWM. It will shock you!

Gigabyte calls the 4th pin on my mobo CPU Opt and Sys 1/2/3 headers as VCC instead of PWM speed control. Many manufacturers are doing this now which is really unfortunate.

I am having am issue with a Swiftech 8 way PWM block

Distinguished

Splendid

Distinguished

Distinguished

Splendid

Distinguished

Splendid

Splendid

Distinguished

Splendid

Distinguished

Distinguished

Splendid

Distinguished

Splendid

Distinguished

Splendid

Distinguished

Splendid

Distinguished

Share this page