Author |
Message |
KnobHell
Joined: Jan 28, 2012 Posts: 56 Location: SLC
|
Posted: Tue May 08, 2012 7:54 am Post subject:
Mankato and SSM2164 negitive rail protection |
|
|
HI,
I have two Mankato's to build soon and I'm wondering if others have used this fix to protect the SS2164 in the case of a negative rail failure?
http://www.milton.arachsys.com/nj71/forums.html?menu=2&submenu=2&subsubmenu=3
In the schematic I can see the series 1n4001 diode, I assume that is not as fast as the BAT85.
len |
|
Back to top
|
|
|
Osal
Joined: Aug 16, 2011 Posts: 147 Location: Here
|
Posted: Tue May 08, 2012 8:21 am Post subject:
|
|
|
To add protection to the negative rail failure and to have reverse polarity protection, just replace the two 1n4001 in the power input of the module by 1A rated Schottky like for example the 1N5819.
http://electronic-sea.net/SSM2164.html _________________ electronic-sea.net |
|
Back to top
|
|
|
KnobHell
Joined: Jan 28, 2012 Posts: 56 Location: SLC
|
Posted: Tue May 08, 2012 8:27 am Post subject:
|
|
|
Perfect! Thank you.
len
Osal wrote: | To add protection to the negative rail failure and to have reverse polarity protection, just replace the two 1n4001 in the power input of the module by 1A rated Schottky like for example the 1N5819.
http://electronic-sea.net/SSM2164.html |
|
|
Back to top
|
|
|
Tim Servo
Joined: Jul 16, 2006 Posts: 924 Location: Silicon Valley
Audio files: 11
|
Posted: Wed May 09, 2012 3:14 pm Post subject:
Mankato and SSM2164 negitive rail protection |
|
|
Yup, changing out the 1N4001s for Schottky diodes on the power input does the trick. Be advised that the failure we're talking about (negative rail going bad, positive rail still okay) is pretty rare, but I guess better safe than sorry.
Tim (Sufferin' Schottkys) Servo |
|
Back to top
|
|
|
Osal
Joined: Aug 16, 2011 Posts: 147 Location: Here
|
Posted: Thu May 10, 2012 1:27 pm Post subject:
|
|
|
Tim Servo wrote: | Yup, changing out the 1N4001s for Schottky diodes on the power input does the trick. |
To replace the 1N4001 by the schotkies, they must be rated as or more than the 1N4001 are. It means that they must be rated 1A or more.
Quote: | Be advised that the failure we're talking about (negative rail going bad, positive rail still okay) is pretty rare, but I guess better safe than sorry. |
This is wrong.
The fact that it didn't happen to you, does not mean that this possibility is pretty rare.
The possibility does exist. Especially in a modular synth DIY context, where the reliability of the power distribution is unknown and where the user interact with it.
If it happens, all the 2164 chips connected to the faulty supply will fail.
We found the cause of the problem and a great solution to it. In modules with reverse protection at the power input, just use rectifier Schotkies instead of 1N4001.
I highly recommend to apply this solution.
http://electronic-sea.net/SSM2164.html _________________ electronic-sea.net |
|
Back to top
|
|
|
Tim Servo
Joined: Jul 16, 2006 Posts: 924 Location: Silicon Valley
Audio files: 11
|
Posted: Thu May 10, 2012 5:57 pm Post subject:
Mankato and SSM2164 negitive rail protection |
|
|
Osal,
I'm not arguing with you, so you can drop the snarky tone. You are right, switching to a Schottky diode (I believe the proper one was specified in an earlier post) does prevent damage to the 2164 chips. It's a neat and very easy fix.
As far as how often this sort of thing happens, I guess selling over 2,000 2164 chips, over 400 Mankato PCBs, and building several Mankato modules myself should count as some sort of experience with this. Oh, and for the record, I designed the Mankato PCB and have personally dealt with every Magic Smoke customer for the last seven years or so. I was surprised to hear about the negative supply failure problem because in ALL that time and with all those customers, I've never had a SINGLE person complain about something even close to this.
Still, you are correct. Changing the 1N4001 diodes in the input protection circuit DOES add an additional layer of protection for the 2164 chip. Again, I'm not arguing with you.
Whatever.
Tim (too tired for this shit today) Servo |
|
Back to top
|
|
|
Sound
Joined: Jun 06, 2006 Posts: 842
Audio files: 1
|
Posted: Fri May 11, 2012 1:07 am Post subject:
|
|
|
|
|
Back to top
|
|
|
Osal
Joined: Aug 16, 2011 Posts: 147 Location: Here
|
Posted: Fri May 11, 2012 1:22 am Post subject:
|
|
|
Tim Servo wrote: | Osal,
I'm not arguing with you, so you can drop the snarky tone. |
I'm not native English. So believe me there is no intention of snarky tone or sarcasm. Just posting my opinion as you do.
Quote: | I've never had a SINGLE person complain about something even close to this. |
It happened to me. I thought you read it in the e-list. All the 2164 chips of my system died due a faulty supply. I wouldn't like that this happen to another person. It is because this, that I recommend to use this protection. Because the possibility of this failure is factual. _________________ electronic-sea.net |
|
Back to top
|
|
|
Neil Johnson
Joined: Aug 11, 2009 Posts: 9 Location: Cambridge, UK
|
Posted: Tue Aug 07, 2012 4:42 am Post subject:
|
|
|
Unfortunately single rail failure on cheap dual-output power supplies is a known failure mode and should be accounted for in the design. More expensive dual-output power supplies have mutual shutdown circuits to make sure that if one rail fails (component failure, regulator over-temp, overload, etc) then all rails turn off to protect the load. This has been common practice on mixer power supplies for years.
If using anti-parallel diodes then swapping them for 1N5817 or other large current Schottky diodes is the right fix. If using series diodes then you need to add another diode as shown on the webpage. As this one is only providing a leakage path for the SSM2164 it can be much smaller, such as a BAT85.
Cheers,
Neil _________________ Reality is as reality does |
|
Back to top
|
|
|
|