EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: paulca on February 27, 2018, 11:50:37 am
-
So I need to replace a PSU in a Chinese product because it has leakage. In particular the FY6600 Sig Gen. I don't like a BNC ground with 120v on it, regardless of the safety or what not.
So I want to replace it with a transformer with linear regulators. The FY6600 has +-12V rails for the output and a 5V rail for it's internals.
I have got part the way there with help from the guys in the FY6600 thread, and am basing it on these components:
Transformer:
https://www.ebay.co.uk/itm/131588122667?_sp=p2488212.m41214.l9765&_trkparms=itemid%3A131588122667 (https://www.ebay.co.uk/itm/131588122667?_sp=p2488212.m41214.l9765&_trkparms=itemid%3A131588122667)
5V regulator:
https://www.ebay.co.uk/itm/271656192244?_sp=p2488212.m41214.l9765&_trkparms=itemid%3A271656192244 (https://www.ebay.co.uk/itm/271656192244?_sp=p2488212.m41214.l9765&_trkparms=itemid%3A271656192244)
+-12V regulator (after adjustment):
https://www.ebay.co.uk/itm/302371834320?_sp=p2488212.m41214.l9765&_trkparms=itemid%3A302371834320 (https://www.ebay.co.uk/itm/302371834320?_sp=p2488212.m41214.l9765&_trkparms=itemid%3A302371834320)
My question is around anything I have forgotten. For example I can think of a few things.
Primary side fuse.
Primary side caps, scrubbers, X, Y caps if I need them.
I googled but I found so many different kinds an different implementations for different countries I figured it would better to ask.
I will happily accept a link to a guide which can be vouched for. I'm in the UK and while the device will never leave my bench I'd it to be safe and as close to "code" as possible.
-
Line frequency leakage current in SMPSUs isn't inherently due to it being a SMPSU, its due to the class Y capacitor from secondary 0V to the negative side of the primary DC bus, that is required as a return path for the HF leakage current capacitively coupled (by the inter-winding capacitance) from the 'hot' end of the primary to the secondary if there isn't an inter-winding screen.
If your proposed linear PSU's transformer doesn't have either a grounded inter-winding screen, or a split bobbin construction, it may still have enough interwinding capacitance to pass enough leakage current to be troublesome, causing its secondary to float at a significant AC voltage.
Unfortunately that transformer on EBAY doesn't look like it has a split bobbin construction and probably doesn't have an inter-winding screen. Unless it has comprehensive manufacturer's data, you'd have to buy it and test it to find its leakage current. If you are floating the secondary, and it doesn't have a proper datasheet, I'd also want to do a 1KV Hi-Pot test between the primary and secondary sides before I'd trust its insulation.
If the leakage current is excessive, the only option is to ground the secondary side 0V rail, which will result in a ground referenced output, though if you need it floating for DC, its possible to use a capacitor between 0V and ground(+ a series resistor to limit the surge if its ever hot-plugged).
If you haven't stated buying parts yet you may want to re-think the PSU design. The +5V rail doesn't need to be extremely quiet so a switching regulator can be used there, with heavy filtering on its output to minimise switching frequency breakthrough,. That means you don't need the 9V windings. I'd also bet you don't need a lot of -12V current, so you *may* be able to capacitively couple a bridge rectifier for the negative rail, so it can all run off a single 15V RMS secondary, though a 15V-0-15V transformer would be preferable.
Have you got figures for the load current required on each rail? If not, I would suggest measuring them using the existing PSU, with both channels analog outputs loaded with 50R 1W resistors (47R would be close enough), while outputting a maximum amplitude squarewave. Don't keep the load on at max output for longer than half a minute (the FY6600 short circuit overload rating).
For the primary feed I'd run the mains through a 1A slow blow fuse in the Line connection, then straight to the primary. If your mains supply is particularly dirty, you may want a suitably rated varistor across the primary. The transformer *should* have a non-resettable thermal fuse buried in the winding. If you add a mains power switch, don't use one with any exposed metalwork.
On the secondary side, use good quality rectifiers rated for at least three times the required output current, and fuse the unregulated rails after the bulk decoupling caps*, with fast-blow fuses rated for the expected maximum output current. To prevent damage to the load if a fuse blows and a rail is lost, connect three diodes between the regulated output rails in order of their voltage,each cathode to the most positive rail, so that if a secondary side fuse blows, the -12V rail cant rise significantly above 0V and the +12V rail cant drop below the 5V rail etc.
* Its possible to fuse between the secondaries and the rectifiers but you'd need four fuses, its *MUCH* more difficult to calculate suitable fuse values, and you'll either have to use slow-blow fuses or go to a significantly higher current rating to handle the initial surge when mains is connected and the bulk decoupling capacitors initially charge.
-
Thanks Ian. A lot to take in.
From the FY6600 I believe the +-12V rails only need to be around 200mA max. Basically enough to power whatever you are driving with the signal generator and in my case right now that would be high impedance audio stuff, but... that could change.
Reports are that he 5V is well under 1A will run off one of the 9V secondaries and no need to parallel the second 9V winding... but it's there, why not?
At least one other user has successfully used this transformer in the FY6600.
I have a spare mains switch, all plastic. I will need to buy a slow blow fuse and holder... or could I get away with the 3 Amp fitted fuse in the UK mains plug of the IEC "kettle" lead?
I also need an IEC 3 pin socket although I not sure about grounding requirements. I would like to leave the output floating if possible so I can connect it to virtual ground devices with elevated virtual grounds without sinking current to earth. (Do I understand that right?)
I'm out of my depth here. AC and Mains... my signature says it all.
-
Paul,
I built a power supply for a similar signal generator (3224) and had boards made to test the supplier. I have 9 left ! If you want a board you are welcome to it and it should sort you out. It is a fully isolated supply using a normal transformer.
Just ignore the SMD components and it needs a 2x15v transformer such as:
http://cpc.farnell.com/camdenboss/ctfc6-15/transformer-6va-2x-15v/dp/TF01334?st=transformer%2015v (http://cpc.farnell.com/camdenboss/ctfc6-15/transformer-6va-2x-15v/dp/TF01334?st=transformer%2015v)
If you want one just PM me.
-
cowasaki, thank you for the kind offer.
I have already ordered the 2x15V 2x9V transformer. I have also ordered the +-15V regulator board and 5V regulator board. That said, I might go the route of making a single board for it. Depends on how things fit or don't fit in the FY6600.
I am going to wait for things to arrive and start testing things step by step.
I have a small enclosure that the transformer while hopefully fit in, so once that is tested to be working and wired as expected, I'll seal it away to keep the mains safe while i work on the rest.
Your board looks nice though :)
-
Place .1uf caps across you diodes in parallel and in parallel across the transformer outputs. It prevents a leaking buzz coming from switching of the diodes which may happen both during small narrow stages of the AC cycle and nowadays additional potential ringing during the time cheap CFL/LED lamps generate EMI on the mains.
Also, yes, get a split bobin transformer for a pre-amp. If you were to go switching, your going to have difficulty getting a medical grade PSU which was designed with the super HI-Pot and no X cap, yet still has approval and is affordable.
See photos of split-bobin transformers:
(https://www.eevblog.com/forum/beginners/my-first-mains-acdc-supply-help/?action=dlattach;attach=399734)
Here are some cheap ones with uni:versal input and output: 1 xformer for +/-12v, a second for +5v, and the last one for +/-15v with a good 350ma to boot.
https://www.digikey.co.uk/product-detail/en/triad-magnetics/FS24-250-C2/237-1502-ND/3986393 (https://www.digikey.co.uk/product-detail/en/triad-magnetics/FS24-250-C2/237-1502-ND/3986393)
https://www.digikey.co.uk/product-detail/en/triad-magnetics/FS12-200-C2/237-1542-ND/3986412 (https://www.digikey.co.uk/product-detail/en/triad-magnetics/FS12-200-C2/237-1542-ND/3986412)
https://www.digikey.co.uk/product-detail/en/triad-magnetics/FS36-350-C2/237-1480-ND/3986352 (https://www.digikey.co.uk/product-detail/en/triad-magnetics/FS36-350-C2/237-1480-ND/3986352)
Here is the complete SplitBobin 4.2kv hipot series. Really well priced and up to 2 amps at +/-18v...
https://www.digikey.co.uk/products/en/transformers/power-transformers/164?FV=ffecd9db%2Cffe000a4&quantity=&ColumnSort=1000011&page=1&pageSize=100 (https://www.digikey.co.uk/products/en/transformers/power-transformers/164?FV=ffecd9db%2Cffe000a4&quantity=&ColumnSort=1000011&page=1&pageSize=100)
Chassis mount split bobin, but only 4kv hipot:
https://www.digikey.co.uk/product-detail/en/triad-magnetics/VPS36-700/237-1288-ND/666174 (https://www.digikey.co.uk/product-detail/en/triad-magnetics/VPS36-700/237-1288-ND/666174)
-
Thanks but I already have an R-Core audio transformer in the post.
-
Also, yes, get a split bobin transformer for a pre-amp.
Ah, you might be assuming this PSU is for my pre-amp project which makes your post make more sense :)
This one is actually for a signal gen to replace a dangerous chinese switching supply with mains leakage. The Audio/Preamp box I will just see if I can filter the PSU better :)
Thanks gain.
-
Thanks but I already have an R-Core audio transformer in the post.
I bet you will have trouble finding that primary/secondary isolation spec...
Just because you see a bobin on the left and right doesn't mean one side is primary and the second side is the secondary.
The transformer ring would be un-balanced power wise. What's nice about the R-Cores is that they are relatively flat for the amount of power they give out. Hence a low profile AV device.
Also, there exist ones with shielding:
-
Thanks but I already have an R-Core audio transformer in the post.
I bet you will have trouble finding that primary/secondary isolation spec...
Just because you see a bobin on the left and right doesn't mean one side is primary and the second side is the secondary.
The transformer ring would be un-balanced power wise. What's nice about the R-Cores is that they are relatively flat for the amount of power they give out. Hence a low profile AV device.
Do I need to know the isolation spec? Assuming I have reports of people using this and it survives 240V, then... do I really care?
As long as it doesn't zap me. I mean is it likely to be worse than a switch mode with a cap between the hot side and device side and 120V on the BNC GND?
-
It does appear to have an inter-winding screen. As long as it's properly implemented, then Earthing that will increase safety and reduce primary-secondary leakage and noise coupling.
-
It does appear to have an inter-winding screen. As long as it's properly implemented, then Earthing that will increase safety and reduce primary-secondary leakage and noise coupling.
When you say "Earthing". Do you mean the SCN connection? I was hoping that was "Screen". Or do you mean just earthing the transformer chassis?
I'd rather like to avoid earth referencing the secondary side and thus my signal generator.
-
Thanks but I already have an R-Core audio transformer in the post.
I bet you will have trouble finding that primary/secondary isolation spec...
Just because you see a bobin on the left and right doesn't mean one side is primary and the second side is the secondary.
The transformer ring would be un-balanced power wise. What's nice about the R-Cores is that they are relatively flat for the amount of power they give out. Hence a low profile AV device.
Do I need to know the isolation spec? Assuming I have reports of people using this and it survives 240V, then... do I really care?
As long as it doesn't zap me. I mean is it likely to be worse than a switch mode with a cap between the hot side and device side and 120V on the BNC GND?
No, obviously the transformer has better isolation than an X cap. I'm sure the transformer is at least 1500v like the one I've attached in this photo. It is safe to touch without feeling a thing. However, I know touching the RCA input on my amp, which uses a toroid transformer, makes a nasty 60Hz hum and I don't feel that current on my finger. On a battery powered amp, without any other connections to anything but the speakers, no hum, if not a very tiny one, not full blase.
The purpose of the HV isolation is so your pre-amp's GND on the RCA ins and outs will sit right there without any current, even electrostatic, bridged from the mains. You want to get as close as possible to a battery powered pre-amp. In fact, you don't need to regulate your +/-15v, even a slight 50/60hz there wont affect your amp's performance since the op-amps reject the power supply pin noise. It's crucial that your GND takes on the reference to what ever device you plug you pre-amp into. Split bobin transformers are known to separate you from the mains as much as possible except for the addition of additional shielding spacing between the windings, or specially insulated windings and layout found in plug in desktop 7-1/2 digit multimeters.
-
It does appear to have an inter-winding screen. As long as it's properly implemented, then Earthing that will increase safety and reduce primary-secondary leakage and noise coupling.
When you say "Earthing". Do you mean the SCN connection? I was hoping that was "Screen". Or do you mean just earthing the transformer chassis?
I'd rather like to avoid earth referencing the secondary side and thus my signal generator.
Yes, the SCN connection. It will (should) be between the primary and secondary winding (insulated from both), so earthing it provides screening/shielding without compromising your floating secondary.
-
For your +/-15v regulator, I like this PCB better than yours, much bigger filter caps, better diodes and it's flat.
https://www.ebay.co.uk/itm/LM317-LM337-Dual-Power-Adjustable-Precision-1-3V-35V-DC-Regulator-Power-Board/142034132784?_trkparms=aid%3D888007%26algo%3DDISC.MBE%26ao%3D1%26asc%3D49925%26meid%3D7f90b9c73a22487f951a6171d7c6c414%26pid%3D100009%26rk%3D1%26rkt%3D1%26sd%3D271656192244%26itm%3D142034132784&_trksid=p2047675.c100009.m1982 (https://www.ebay.co.uk/itm/LM317-LM337-Dual-Power-Adjustable-Precision-1-3V-35V-DC-Regulator-Power-Board/142034132784?_trkparms=aid%3D888007%26algo%3DDISC.MBE%26ao%3D1%26asc%3D49925%26meid%3D7f90b9c73a22487f951a6171d7c6c414%26pid%3D100009%26rk%3D1%26rkt%3D1%26sd%3D271656192244%26itm%3D142034132784&_trksid=p2047675.c100009.m1982)
This guy already has the 0.1uf caps across the diodes.....
-
Looks nice too.
As the xformer will take forever and a day to get here, I ordered a cheap xformer from RS which should arrive Friday. It's +-18V sealed PCB mount, 4W. Also got some -15 +15 regs, and Nichicon 1000uF caps. The rest I think I have in stock. It will mean I can have a play transformer and experiment which not blowing shit up.
Was reading a marketing PDF for some chinese transformers and something lost in translation which made me laugh was the line, "Better explosive voltage than other transformers". Not sure that's what they meant LOL.
-
I'd suggest reading the LM317 datasheet. There may even be an old application note from when they first came out. I built a 5 V supply for TTL work based on that general line of regulators and used an LM317 to fix a burned out 12 V CB radio PSU. I think I modified the latter to include an extra tantalum capacitor to reduce ripple that was suggested. But it was a very long time ago.
-
Looks nice too.
As the xformer will take forever and a day to get here, I ordered a cheap xformer from RS which should arrive Friday. It's +-18V sealed PCB mount, 4W. Also got some -15 +15 regs, and Nichicon 1000uF caps. The rest I think I have in stock. It will mean I can have a play transformer and experiment which not blowing shit up.
Get some heat sinks for your regulators! 18 Vrms is ~25 V peak, which will be approximately the input to your regulator after the smoothing caps. (Simulate to see!) You’ll drop 10 V in the regulator which is a watt with a modest 100 mA load.
-
Here is the video - https://www.youtube.com/watch?v=CIGjActDeoM (https://www.youtube.com/watch?v=CIGjActDeoM)
-
I did watch that. I'm not really after creating a fully variable bench supply. The goal here is to play with transformers, make a fixed dual rail supply and when I get that working I can build the supply for the FY6600 to replace the dodgy crap it comes with.
-
So I got my "play" transformer through from RS. No regulators yet, but I put the transformer onto some strip board with a 250V switch and a 0.1A fuse primary side. Shrink wrapped and electrical taped up the primary side to make it at least resembling "safe".
Power it on, no smoke, no bangs.
Measured the voltage on the secondary in AC Trms mode. It reads 25V on each output. Fine.
It's supposed to be a 2x18V transformer and while I understand it will read high without a load and even putting it through a bridge rectifier and a cap will drop it considerably before a regulator can bring it down to 15V...
So the question is... can I connect my little mini bench scope to it. It has a 50V pk printed on the BNC. So if that's 50V peak it's fine as 25Vrms should be 35V peak. If it's 50V peak to peak it's not fine.
I have my OWON scope which has a much higher input rating, but it needs the laptop which is elsewhere tonight.
-
Actually, I think I'm okay. It's 50V peak with 1x probe. So it should be fine with 70Vpp with a 10x probe. 1MOhm input impedance.
-
It's supposed to be a 2x18V transformer and while I understand it will read high without a load and even putting it through a bridge rectifier and a cap will drop it considerably before a regulator can bring it down to 15V...
No, putting it through a bridge rectifier and cap will bring it up to the peak voltage minus diode drops. Your regulator best be comfy with seeing up to 40V unloaded.
-
It's supposed to be a 2x18V transformer and while I understand it will read high without a load and even putting it through a bridge rectifier and a cap will drop it considerably before a regulator can bring it down to 15V...
No, putting it through a bridge rectifier and cap will bring it up to the peak voltage minus diode drops. Your regulator best be comfy with seeing up to 40V unloaded.
Measured it was 7Vpp at 10x, so that's 70Vpp. Through a bridge rectifier will it lose one or two diode drops? Anyway, worst case is it loses one, so it's still 34.3V peak. Positive regulator 7815 is good for 35V, negative 7915 is good for -40V. I know it's close, but it should be fine. The regulators themselves will put 3.5mA + and 8mA - load on it which will bring the voltage down I expect to something well under the regs ratings.
I assume it also means though that the secondary side is worth being careful around too.
-
Two drops.
The secondary side is always worth being careful around. The voltages aren't dangerous, but you're presumably not dealing with a current limited (yes, those exist - they're really low power to begin with) transformer - which means shorting it will bring smoke, fire, and sadness if the fuses aren't sized right. Don't assume a primary side fuse can protect the secondary effectively!
-
Transformer is one of these:
https://docs-emea.rs-online.com/webdocs/0f3c/0900766b80f3c2a1.pdf
The 2x18V 100mA (each) variant.
No it's not limited and it does mention powering a DC load off it will not cater for the full 100mA DC. The then tell you to phone their customer services to find out LOL. I assume it can be calculated. Besides I don't think I will load it anywhere near to 100mA, I don't intend to power much off it, maybe a headphone amp to see if it's smoother than a switch mode one.
Remember this is a test dummy to play with while I wait on the RCore from China.
EDIT: For a basic "play" rectifier, can I use basic 1N400x diodes? 1N4002 or higher reverse voltage rating?
-
Not sure if you looked but theirs a few videos on the FY3224S Sig Gen.Granted they may not be the same supply but it might give you some insight on the subject.
https://www.youtube.com/watch?v=ML-lmuHoh-0 (https://www.youtube.com/watch?v=ML-lmuHoh-0)
also one on USB isolation
https://www.youtube.com/watch?v=HqF_1y3U_qg (https://www.youtube.com/watch?v=HqF_1y3U_qg)
-
EDIT: For a basic "play" rectifier, can I use basic 1N400x diodes? 1N4002 or higher reverse voltage rating?
There's exactly nothing wrong with a bridge rectifier made out of 1N400x diodes. 1N4001s are fine for this voltage.
-
It's supposed to be a 2x18V transformer and while I understand it will read high without a load and even putting it through a bridge rectifier and a cap will drop it considerably before a regulator can bring it down to 15V...
No, putting it through a bridge rectifier and cap will bring it up to the peak voltage minus diode drops. Your regulator best be comfy with seeing up to 40V unloaded.
Measured it was 7Vpp at 10x, so that's 70Vpp. Through a bridge rectifier will it lose one or two diode drops? Anyway, worst case is it loses one, so it's still 34.3V peak. Positive regulator 7815 is good for 35V, negative 7915 is good for -40V. I know it's close, but it should be fine. The regulators themselves will put 3.5mA + and 8mA - load on it which will bring the voltage down I expect to something well under the regs ratings.
I assume it also means though that the secondary side is worth being careful around too.
It seems your transformer has a center tape, which means you will be using FULL BRIDGE RECTIFIERS with 4 diodes - in this case, it is two diode drops, which may be ~1.2V (if Si) or ~0.4V (if Schottky).
Regarding the input voltages to the regulators, be careful if the mains voltage goes too much above what you measured, as the regulators may operate above their specs. Sometimes simply adding some bleeder resistors on the tank capacitors before the regulators can create enough load to reduce the transformer's output voltage.
-
So, it functions. A first prototype anyway.
Luckily I spotted that the 1000uF caps I bought were only 25V before pushing 34.4V through them.
Managed to find some 50V ones, but they were only 100uF. Anyway they (2 in parallel) did fine to run an LED with only 50mVpp ripple. The 25V ones will be fine on a 9V transformer to run the 5V rail on the FY6600 supply.
The bare transformer tested:
(http://i.imgur.com/Je4fZPJ.jpg) (https://imgur.com/Je4fZPJ)
With bridge rectifier, caps and 15V regulator + test dummy LED with 1k resistor.
(http://i.imgur.com/Uk2vyLc.jpg?1) (https://imgur.com/Uk2vyLc)
-
So, rough and ready schematic for the FY6600 PSU.
The question is, am I on the right track? I'm fairly sure it can be simplified a bit. I am showing two transformers when it will really be one that provides 4 secondaries.
(http://i.imgur.com/PLStKhy.png) (https://imgur.com/PLStKhy)
-
Your schematics look alright. Just check the datasheet of your regulators to see if they require specific capacitors in the input/output for stability.
For example, you mention LM7805, LM7812 and LM7912:
http://www.ti.com/lit/ds/symlink/lm7800.pdf (http://www.ti.com/lit/ds/symlink/lm7800.pdf)
http://www.ti.com/lit/ds/symlink/lm79.pdf (http://www.ti.com/lit/ds/symlink/lm79.pdf)
If you want to minimize the ripple with what you have, simply put two 1000uF/25V capacitors in series and you will have a 500uF/50V capacitor.
http://farside.ph.utexas.edu/teaching/302l/lectures/node46.html (http://farside.ph.utexas.edu/teaching/302l/lectures/node46.html)
-
If you want to minimize the ripple with what you have, simply put two 1000uF/25V capacitors in series and you will have a 500uF/50V capacitor.
http://farside.ph.utexas.edu/teaching/302l/lectures/node46.html (http://farside.ph.utexas.edu/teaching/302l/lectures/node46.html)
I googled that and found while it was possible I needed to put meg resistors in parallel to ensure the caps don't charge to different voltages, but maybe they are being pedantic.
I decided to try it with just the 100uF 50V caps in parallel, giving 200uF. I might try the 1000uF 25s in series with 1 meg resistors across each, just takes up more breadboard room.
I'll order 63V 1000uFs for the final version. (Edit: I'll calculate what capacitance I need for a guesstimated load and aim a little higher)
Toying with making a strip board version, but laying out a PCB anyway if my first 2 boards actually arrive without too many customs charges that is.
The FY6600 hasn't arrived yet though, it should be here soon.
-
The +/-12V supply only needs one bridge rectifier (although it should have a minimum 70V rating). Connect the 15V RMS windings in series, use the resulting center tap as 0V, and connect the bridge AC terminals to the ends of the two windings. The bridge + and - terminals will provide the raw DC + and - rails.
Connect a 10nF capacitor across each diode in each bridge rectifier as close to the bridge rectifier as possible to suppress impulse noise due to rapid current cut-off during reverse recovery.
Connect 100nF ceramics as close to each regulator as possible in parallel with their input and output decoupling capacitors.
-
I only see a huge benefit in using equalizing 1M $\Omega\$ resistors in more extreme cases - i.e., not with the voltage and current levels you are using, but others may disagree.
At any rate, the series association was just a suggestion for you to experiment and play around with the circuit.
-
The +/-12V supply only needs one bridge rectifier (although it should have a minimum 70V rating).
Gosh, how did I miss that? You are right (I need more coffee).
(edit) Something like:
(http://static.elitesecurity.org/uploads/2/4/2459916/basic_symmetric_power_supply.gif)
-
I did think about that, but wasn't sure how to get the common 0V rail or how the rectifier would be wired.
When you say "the centre tap", if they are two separate secondaries, I assume I just take the point I put them in series, the common point?
-
I did think about that, but wasn't sure how to get the common 0V rail or how the rectifier would be wired.
When you say "the centre tap", if they are two separate secondaries, I assume I just take the point I put them in series, the common point?
You need to find out the "polarity" of the windings, or you will end up with zero volts.
To find the polarity of the windings you can use your oscilloscope: tie two wires together and measure the output of the two disconnected wires - if it is close to zero volts, you need to disconnect and tie the other wire of one of the windings.
You could also check with an analog multimeter or a digital one with a good bargraph.
https://www.eevblog.com/forum/beginners/analog-meter-vs-a-dmm/msg210926/#msg210926 (https://www.eevblog.com/forum/beginners/analog-meter-vs-a-dmm/msg210926/#msg210926)
-
Yes. the junction of the two secondaries is the center tap.
As RJSouza just posted, be careful about phasing - if in doubt connect one end each of a pair of secondaries and check the voltage across the remaining ends before wiring them up any further - its especially critical for your paralleled secondaries for the 5V supply. For the +/-12V supply, it wont instantly destroy the transformer, but getting it wrong doubles the ripple and stresses the transformer more for the same current. Any multimeter on AC V should give a clear indication, I wouldn't bother scoping it. Don't expect *exactly* zero volts between the ends you are going to parallel - if its down around 1% of the nominal secondary voltage you shouldn't have any issues.
-
The 7912 I used required a tiny load in order to regulate the voltage. I created that load by using the -12v rail for the power LED. At least that way you can test it without connecting it.
-
Hate to say this but those transformers are a bit to high.You have to calculate the voltage after the filter caps.You do this by taking say your 15 volt transformer and multiplying 1.4 that will give you the voltage that will be going to the regulator .You'll end up with around 20 volts off your 12 volt transformer and 12.6 of the 9 volt transformer.Remember the more voltage you put to the regulator the hotter it will get .So typically you want to shoot for about 4 -5 volts above the required voltage.So a 12 volt transformer on the 12 volt side will get you around 16-17 volt after rectifier and filter cap.And I would use a 6 volt on the 5 volt side which would be around 8 volts after rectification and filtering.You also don't want to go to high on the filter cap either .Because on startup to high uF can put to much load on the transformer shortening it's life and it will blow fuses.Capacitor at zero charge acts like a short circuit before they begin to charge.
-
Its *FAR* more complex than that. You *MUST* allow for the ripple on the unregulated rail as the regulator input needs tp stay above its dropout voltage at the maximum load current, to avoid loss of regulation and excessive output ripple.
The reservoir capacitor charges up to the peak of the full-wave rectified AC (less diode losses) and then has to supply the load current until the next full-wave peak, 10ms later. The discharge slope can be calculated from Q=CV. You can approximate the ripple by Vripple=I/(2fC), (where f is the mains frequency) though for a more accurate solution you have to account for the discharge phase ending earlier than the peak and waveform distortion due to the transformer source impedance and the diodes dynamic resistance, which makes a pure analytic solution rather awkward. Its easiest to simply simulate it to check out the effect of load current and capacitance value.
A LTspice sim (attached) shows that for 9V RMS in (loaded), 1000uF of reservoir capacitance, and 1A silicon diodes, you wont even be able to draw 600mA without the ripple trough dropping below 7V. You also have to worry about the ripple current rating of the reservoir capacitor - the sim shows 1A RMS ripple current, which is a *LOT* for a 1000uF capacitor, so it will cook itself to death. More capacitance usually gets you a higher ripple current rating, (especially if you use several smaller capacitors in parallel to increase their surface area and thus their heat dissipation capability), but as JWillis has pointed out, that can get you in trouble with the startup surge current so you'll probably have to uprate the diodes and use a slow-blow fuse.
You also have to consider the effects of mains voltage variation - can your design maintain the ripple trough voltage above the regulator dropout voltage with the mains voltage at the permitted lower limit, and will the peak input voltage, with no load and the mains voltage at its permitted upper limit, exceed the regulator's maximum input voltage?
(see http://www.twothirtyvolts.org.uk/pdfs/site-info/Explanation_230Volts.pdf (http://www.twothirtyvolts.org.uk/pdfs/site-info/Explanation_230Volts.pdf) )
-
Thanks. So I have some things to consider.
On the load, I believe the expected load is lower than 1A. I haven't seen any true characteristics of the FY6600, but in bench tests it was only something like 250mA on the 12V rails and not much more than that on the 5V rail.
As long as I don't connect a low impedance device to the BNCs....
The video posted, I believe took those measured loads and calculated the cap values based on that, but they had a much lower 5V transformer rated at 6V, so they had to use a 400mV LDO regulator. They ended up with (I believe from memory) 2200uF caps.
As my xformer will be putting out higher voltage I shouldn't need the caps to be that large. Though I might swap the 1A fast blow for a 0.5A slow blow.
So plenty of things to work out. I have basic TO-220 heatsinks and thermal grease, but I'll need to work out the actual heat dissipation of the regulators due to the high input voltage. I have the options of buying (or hacking out of something) a much heavier heatsink and/or fitting a 3cm 5V fan to the back of the case. Another option is making a heatisnk, I have a 3" by 2" sheet of what looks like thick aluminium which was the heat sink from a power bank battery bay, I could bend it in some corigated fashion and mount all 3 regulators to it.
Note, this device will not be an always on type thing. It will come out once in a while, run for an hour and go back on the shelf. So it will be full supervised and I'm hoping the worst it can do is fry the FY6600, blow it's fuses, let out some smoke and pop my RCD. Oh... of course it could potentially dump 120V/240V onto the BNC and fry me and anything connected, but that seems unlikely. Of course I'd like to do what I can to avoid all of the above.
-
If in the LTspice sim, you put a voltage source equal to the output voltage in series with the current source that is the load (below it is best), the load's dissipation will be almost exactly equal to that of a LM7xx regulator passing the same current (neglecting the regulator's quiescent current). You can then alt-click the current source to plot its dissipation, and ctrl-click the resulting trace legend to calculate the average power.
Of course, that only helps if you know the FY6600's current consumption on each rail - but you can measure that with the existing PSU, and your new transformer's output voltage when loaded with a RMS current of 1.6 times the DC current, or with the actual DC current after the bridge + reservoir capacitor.
Taking the example of your nominally 18V RMS @100mA transformer, its nominally 230V in so at 240V the output voltage will already be 18.8V, nearly 5% high. You measured its no-load voltage as 25V. At 80mA RMS, for 50mA DC out, I estimate the voltage (by interpolation) will be 20V RMS. Plugging that into the sim in place of the 9V the source Vac was set to, changing the cap to 100uF, decreasing Iload to 50mA and adding a 12V source so the regulator dissipation can be plotted, I get 645mW for the regulator dissipation. That does *NOT* include the power dissipation due the the quiescent current (from In to Gnd), which is another 126mW, so the total LM7812 dissipation at 50mA load current with 20V RMS in will be 0.77W. Revised sim for attached.
Add up the expected regulator dissipations for all the rails, and the result should be close enough to choose a heatsink size. I strongly suspect you'll need an external heatsink, or fan cooled as I expect you'll have a worst-case dissipation of not much under 10W
Don't put any fan on the 5V rail as it makes the regulator dissipation problem worse. The full wave rectified 9V rail should be in the right ballpark to run a 12V fan a bit slower than normal.
-
Thanks Ian.
Remember the test dummy xformer is 18V regulated to 15V, the FY6600 xformer is only 15V regulated to 12V. Thought that might result in similar heat disapation.
So if an 18V xformer gives me 25Vrms / 35Vpk the 15V might be around 20.5Vrms (based on a extrapolated 29Vpk). That's unloaded.
Your spice sim won't run for me, simulate and most other things are greyed out, but my LTSpice currently has issues in Linux, might work natively in windows if I reboot.
The xformer arrived, still no FY6600. So I could end up with a rather nice audio PSU and no sig gen :) Nor has the +-12V PSU kit board I ordered.
I haven't powered it up, maybe over the weekend, it looks fairly well done. The primary and secondard windings are around what looks like high temp plastic or ceramic bobbins which wrap each side of the core. The chassis protrudes up the middle, but not fully. So the windings do have separation. It's hard to see but it's not much, maybe a millimeter. That might suggest it would not have a high isolation value and could potentially arc across with a HV surge. But I'm grasping in the dark there.
Winding outputs are not polarized in any way. Same colour wire both ends, so I will need to use the meter to find the polarity for commoning.
Ordered some 63V 1000uF and 470uF caps. If I need to replace the capcitors in the +-12V and +5 regulator boards. They will also allow me to make the test xformer circuit simpler and test commoning it's two 18V secondaries.
Out of interest, my mains measures at 239V (from memory), been a while since I tested it.
-
So I got a pair of regulator boards. An LM317/LM337 +-12V (adjustable) AC-DC regulator and am LM7805 5V regulator AC/DC regulator.
I want to test the +-12V regulator, however, using the "test" transformer as I have not wired the other one up yet. It has two 18V windings which unloaded put out 35V.
So I checked the polarity of the windings and they were as the pinout on the transformer would intuitively lead you to. Swapped the bridge connection and got 0 volts on the scope to confirm and went back to the correct polarity.
The LM317 board is rated for +-35V and takes the AC input as +, - and 0. I was assuming I would connect the outer windings to + and - and the centre point to 0.
The scope however is making me wonder. When measuring the voltage across both windings I was getting....
VMax: 70V
Vpp: 115V
This sounds like twice what it should be. Is this because I was measuring the outer windings? Would that not mean while one winding is+35V the other winding is -35V, giving 70V output? it The 115V is because the bottom of the sinewave was squashed. Not entirely sure why.
Obviously when connected to the regulator board the + side will see 0-35V and the - side will see 35-0V
It's just confusing trying to work with + and - terms with AC. I'm confuddled.
EDIT: The reason I'm being overly cautious is the the caps on the board are only rated to 35V and I don't want one popping in my face. I also intend to put some form of load on it to hopefully bring the voltage down a bit.
-
Oh dear. 35V unloaded is right on the bleeding edge for those caps. If you want a long and happy life from them, replace with 50V or even 63V rated ones. The LM317 and LM337 are only rated for 40V Abs. Max. voltage differential in to out so there's *very* little margin for supply voltage variation at startup. OTH. once its powered up, there's 15V less across each regulator as long as their outputs never get shorted so they should be OK. Do re-check the measurements and this time check the mains supply voltage as you want to be certain it wasn't significantly under the UK's customary 240V at the time of measurement.
Worst case, it may benefit from a capacitance multiplier circuit on each of the raw + and - rails, which will vastly reduce the ripple 'seen' by the regulators, and if you clamp the multiplier transistors' bases with 27V Zeners, will also act as a preregulator when it is lightly loaded.
-
It shouldn't be a big deal, the RCore has 15V windings rather than 18V. Maybe I should wire it up instead. I was just going to test with this 18V xformer as it's wired and I have it handy.
I have got Nichion 63V 1000uF caps, I could swap them, better quality and they would be well overrated too. Probably means taking the risk of drilling the holes back out as I can never get them clear after removals.
The high reading on the scope, would this have been because I was in AC coupling mode while measuring across two AC windings?
Worst case, it may benefit from a capacitance multiplier circuit on each of the raw + and - rails, which will vastly reduce the ripple 'seen' by the regulators, and if you clamp the multiplier transistors' bases with 27V Zeners, will also act as a preregulator when it is lightly loaded.
I assume when the power supply is in it's new home powering the device it will not spend much time unloaded. Of course it depends on the output mosfets on the sig gens outputs as to how unloaded the +-12V rails are.
The SigGen is still on the slow boat from China.