EEVblog Electronics Community Forum
General => General Technical Chat => Topic started by: n_s_fletcher on February 26, 2012, 03:12:15 pm
-
Hi,
I just found this forum via Google, and get the impression that there are one or two folks here who have had issues with Farnell in the past...
I've just built a fixed output +/-12V, +/-15V PSU for a hobby project that I'm working on. The positive side worked like a dream, but I've had real fun and games getting the negative side to work.
I already had +/-12V regulators, but needed some +/-15V regulators in a hurry so decided to buy them from Maplin. Anyway, started to assemble the circuit, 78 series regulators have a standard pin-out of: Input, Common (Gnd), Output. I haven't used 79 series regulators in a while so checked a datasheet that I had on my PC, the pin-out is: Common (Gnd), Input, Output. I assume this is a standardised pin-out?
So I finish building the circuit, re-check the layout, check for shorts etc, and go for switch-on... The positive side is ok, I get + 12 and + 15V no problem. However, on the negative side I get -19V on both the -12V and -15V outputs! BTW both these regulators were made by ST Microelectronics.
Ok, so I removed both regulators from the circuit and tried them again in plug board on the bench, still getting -19V, so I assumed that I must have blown them up somehow. I had a rummage in my junk box, and found an ancient uA7912 (Fairchild?) regulator. Checked the pin-out, put it into my circuit and hey presto I get -12V!
So I decided to order a selection of 78 and 79 series regulators from Farnell, to replace the one’s I had apparently blown up. I deliberately downloaded the datasheets from Farnell’s website for the regulators I bought, so there is now no confusion about the pin-out for these devices.
They arrive last week, and on Friday afternoon I start to fit the new regulators into my circuit, making sure they are wired up as per the datasheets. This time the 7915 regulator worked fine, giving out -15V, this regulator was manufactured by ON semiconductors. However, the 7912 regulators I bought, manufactured by ST Microelectronics, still just gives me -19V! So I stuck the old Fairchild uA7912 back in there, and we get -12V. So I think I can safely say that this isn’t the circuit layout or device pin-out that’s the issue here, leaving the only other possible explanation being that these 7912 regulators are faulty.
http://uk.farnell.com/stmicroelectronics/l7912acv/ic-v-reg-12v-7912-to-220-3/dp/1095391?Ntt=1095391 (http://uk.farnell.com/stmicroelectronics/l7912acv/ic-v-reg-12v-7912-to-220-3/dp/1095391?Ntt=1095391)
Has anyone else been having problems with these regulators?
-
Did you have a minimum load on the regulators when you measured the voltage? The 78xx and 79xx require a load of at least 5 mA, if I remember correctly. If you measure their output without a load there's no telling what voltage they will output.
-
I haven't used 79 series regulators in a while so checked a datasheet that I had on my PC, the pin-out is: Common (Gnd), Input, Output. I assume this is a standardised pin-out?
Yes, but when in doubt, always check the original (in this case ST) datasheet. Note that the tab is also connected to the input terminal, so make sure it's isolated.
They arrive last week, and on Friday afternoon I start to fit the new regulators into my circuit, making sure they are wired up as per the datasheets. This time the 7915 regulator worked fine, giving out -15V, this regulator was manufactured by ON semiconductors. However, the 7912 regulators I bought, manufactured by ST Microelectronics, still just gives me -19V! So I stuck the old Fairchild uA7912 back in there, and we get -12V. So I think I can safely say that this isn’t the circuit layout or device pin-out that’s the issue here, leaving the only other possible explanation being that these 7912 regulators are faulty.
Or that the ST regulators are more sensitive than the Fairchild ones. Did you follow the datasheet recommendations? The output cap is not optional for the 79xx, unlike the 78xx. Do you have at least 1 uF tantalum electrolytic or 25 uF aluminum elecrolytic close to the output? Do you have an input cap? If you have excessive output capacitance, do you have an anti-parallel diode? It seems the footnote symbols got cut from the ST datasheet, so double check with the National datasheet.
Do you measure close to zero ohms resistance between input and output? Remove the regulator, still zero ohms between the nets in the circuit -> circuit error. Zero ohms between the pins of the regulator without anything else (including heat sink) connected -> dead regulator.
If you're using those horrible solderless breadboards, triple check the connections (measure voltages both on the pin and on the breadboard), they are not designed for the rather large TO-220 legs.
-
If what your saying is true, then the load regulation of these devices, in general, must be extremely poor, which is not my experience.
-
What does the datasheet say about load regulation below 5 mA? All those tables are not just there to look pretty, they actually contain some important information. Voltage regulation is also quite poor when Vin is below for a 7912 below 13.5 V or so, does that mean that voltage regulation in general is very poor?
A common way to get around the minimum load requirements is to connect a resistor or LED across the output.
-
I haven't used 79 series regulators in a while so checked a datasheet that I had on my PC, the pin-out is: Common (Gnd), Input, Output. I assume this is a standardised pin-out?
Yes, but when in doubt, always check the original (in this case ST) datasheet. Note that the tab is also connected to the input terminal, so make sure it's isolated.
Hi Alm, Yep, I've done that. The Farnell datasheet is an ST original.
They arrive last week, and on Friday afternoon I start to fit the new regulators into my circuit, making sure they are wired up as per the datasheets. This time the 7915 regulator worked fine, giving out -15V, this regulator was manufactured by ON semiconductors. However, the 7912 regulators I bought, manufactured by ST Microelectronics, still just gives me -19V! So I stuck the old Fairchild uA7912 back in there, and we get -12V. So I think I can safely say that this isn’t the circuit layout or device pin-out that’s the issue here, leaving the only other possible explanation being that these 7912 regulators are faulty.
Or that the ST regulators are more sensitive than the Fairchild ones. Did you follow the datasheet recommendations? The output cap is not optional for the 79xx, unlike the 78xx. Do you have at least 1 uF tantalum electrolytic or 25 uF aluminum elecrolytic close to the output? Do you have an input cap? If you have excessive output capacitance, do you have an anti-parallel diode? It seems the footnote symbols got cut from the ST datasheet, so double check with the National datasheet.
Do you measure close to zero ohms resistance between input and output? Remove the regulator, still zero ohms between the nets in the circuit -> circuit error. Zero ohms between the pins of the regulator without anything else (including heat sink) connected -> dead regulator.
I've been using 100uF electrolytics on the output. I don't use anti-parallel diodes, as I can't really see the point. Silly mistakes have already been ruled out (I hope!) ;D
If you're using those horrible solderless breadboards, triple check the connections (measure voltages both on the pin and on the breadboard), they are not designed for the rather large TO-220 legs.
Good point, but I don't think it's a factor here. One thing I do agree with you on is that there is something really weird going on with these regulators. Will get another one out the packet and try a 1uF cap plus a load, a resistor plus LED, and let you know how I get on.
-
Problems with ST negative regulators have been discussed in other threads of the Forum.
https://www.eevblog.com/forum/projects-designs-and-technical-stuff/lm7905-noise-issues-(stmicroelectronics)/msg81804/#msg81804 (https://www.eevblog.com/forum/projects-designs-and-technical-stuff/lm7905-noise-issues-(stmicroelectronics)/msg81804/#msg81804)
My experience says that it is better to avoid them, because of the weird, unstable results you get.
Best regards
-
Problems with ST negative regulators have been discussed in other threads of the Forum.
https://www.eevblog.com/forum/projects-designs-and-technical-stuff/lm7905-noise-issues-(stmicroelectronics)/msg81804/#msg81804 (https://www.eevblog.com/forum/projects-designs-and-technical-stuff/lm7905-noise-issues-(stmicroelectronics)/msg81804/#msg81804)
My experience says that it is better to avoid them, because of the weird, unstable results you get.
Best regards
Thanks for the link Ciccio :o
Well, I have tried the two remaining regulators I have left with a 100uF cap and a resistor and LED as a load... and they give out -12V ???
The other three still don't work...
The only conclusion I can come to is that you need to have a minimum load on the output at all times, otherwise they blow up! :o
These ST Microelectronics regulators do seem to give weird results, so I tend to agree that they are best avoided.
With regards to the oscillation problem mentioned in the link, I experienced a similar problem in an op-amp circuit being used to drive a motor. ST Microelectronics 7812 and 7912 regulators were being used to power the circuit, but it was the 7812 regulator which was becoming unstable, and it's output voltage was dropping below the minimum output level. Changing the output caps on both regulators from 1uF tants to 100uF electrolytics, and fitting an 100nF capacitor between the input and output terminals on the 7812 seemed to cure the problem for good. What I have learned today seems to confirm what I thought at the time that the problem was being caused, at least in part, by the voltage regulators.
Anyway, thanks for your help guys :)