Replaced about 10 fluro light fittings the other day, these had some exceptional looking ballasts in them considering the fittings were over 30 years old and on all day.
Quality made in Australia ballasts, I thought could I use these as cokes in power supplies? has anyone tried this? they would fit well under a conventional chassis.
Quality made in Australia ballasts, I thought could I use these as cokes in power supplies? has anyone tried this? they would fit well under a conventional chassis.
An externally hosted image should be here but it was not working when we last tested it.
Slip of tongue? (OK, they're white, understandable)
Discussed here for example:
http://www.diyaudio.com/forums/power-supplies/177117-fluorecent-choke-power-supply.html
I've measured a few flouro chokes, and the 18W 240VAC application type seem the most commonly found with the highest inductance, although you may be able to come across lower 'power' application chokes (eg. 9W) which should have higher inductance levels. The inductance droop with DC current is quite sedate, and applying even a cautious power dissipation of say 5W to the choke allows use in almost any amp as far as DC current rating goes. The choke construction is also pretty much designed for large applied AC voltage and high voltage insulation to core, so little cause for concern there. Example measurement in link:
http://dalmura.com.au/projects/Choke measurement.pdf
I've used them in about three renovated old PA amps where the original capacitor input rectifier gave a quite high B+ with substantial ripple voltage. In that situation I added a CL stage after the rectifier - the C is a lowish value 380VAC poly capacitor (eg. 2.2-5uF) and with the flouro choke that then provided a great per-filter to the main B+ filter cap (ie. lower B+ voltage and ripple level and the B+ electrolytic cap has lower ripple stress). Most old PA amps have room to fit these parts 🙂
http://dalmura.com.au/projects/Choke measurement.pdf
I've used them in about three renovated old PA amps where the original capacitor input rectifier gave a quite high B+ with substantial ripple voltage. In that situation I added a CL stage after the rectifier - the C is a lowish value 380VAC poly capacitor (eg. 2.2-5uF) and with the flouro choke that then provided a great per-filter to the main B+ filter cap (ie. lower B+ voltage and ripple level and the B+ electrolytic cap has lower ripple stress). Most old PA amps have room to fit these parts 🙂
The Australian "Mudlark" amplifiers use fluoro ballasts in their parafeed circuits - as output tube plate loads.
No no no put them back where you got it from!
Electronic ballasts transmitt broadband interference all over the HF bands...😱
Electronic ballasts transmitt broadband interference all over the HF bands...😱
My personal experience leads me to say that I have seen nothing that suggests this to be true when used as a DC choke. I have now used them in two preamps and two power amps without issue.
Shoog
I mean when using /electronic/ ballasts (not magnetic) in fluorescent lights.
The OP has robbed a couple of fluroescent lights of their precious EMI-FREE magnetic ballasts...
I was half-joking. Seriously though the more electronic ballasts in the world the worse off it will be for all of us, EMI wise.
Has the penny dropped yet? 😀
The EC9 type 9W flouro ballast is still available new at about AUD$6 - damn cheap, and very compact. It appears that this is the ~3-4H choke used in the Mudlark
As they are not designed to carry DC, they have no proper engineered airgap. However, these chokes were made as cheap as possible and generally do not have proper interleaving of the laminations. Plus the two lamination stacks are welded together, which locally destroys much of the magnetic properties. So, from these two factors, they have a sort of fortuitous magnetic gap anyway.
You may be lucky and they work ok. But I would expect that in power supply service they are only somewhat better than a resistor having the same DC resistance the choke has when hot. Not anywhere as good as a choke designed for the purpose of reducing power supply ripple.
Those that think that a flouro ballast worked well in their amp should try using a resistor instead. They might find the resistor can give much the same results in suppressing ripple.
Another, and perhaps more important factor is this: Because they are built in a low profile long shape to fit in the flouro housing, their leakage magnetic field is immense compared to normal power supply chokes. Therefore expect a lot more trouble with hum getting into signal circuits thru magnetic coupling.
You may be lucky and they work ok. But I would expect that in power supply service they are only somewhat better than a resistor having the same DC resistance the choke has when hot. Not anywhere as good as a choke designed for the purpose of reducing power supply ripple.
Those that think that a flouro ballast worked well in their amp should try using a resistor instead. They might find the resistor can give much the same results in suppressing ripple.
Another, and perhaps more important factor is this: Because they are built in a low profile long shape to fit in the flouro housing, their leakage magnetic field is immense compared to normal power supply chokes. Therefore expect a lot more trouble with hum getting into signal circuits thru magnetic coupling.
Last edited:
There seem to be many people for and against using these, magnetic leakage would be troublesome, I would have to design around that. I guess the best way is to try some out, I have a couple of 9W units also about half the physical size of the 18W......
What about the "electronic chokes" for fluros?
Can they be used?
See http://www.diyaudio.com/forums/powe...ctronic-choke.html?highlight=electronic+choke
Kevin
Can they be used?
See http://www.diyaudio.com/forums/powe...ctronic-choke.html?highlight=electronic+choke
Kevin
Last edited:
Is there something I'm missing? why all this interest in ELECTRONIC chokes?????
Hello,
In addition to the concerns raised by the other gentlemen, you should consider this: a decent audio power supply choke will be constructed to minimise the self-capacitance between the windings, its flux leakage and susceptibility. This is significant, because the self-cap is in parallel with the inductance you want to use to reduce the power supply noise etc. This "cap" will do several bad things- it will form a parallel resonant circuit at several frequencies- quite a bit more complicated than the basic text book stuff.
Test this for yourself: connect a good low impedance power supply (12V DC is fine) to a simple Nfet power stage. Play safe - fuses etc-- 1 R in source lead, 10K from gate to source, bias the gate at 0 V with a 100K pot across the rails and a 10 K resistor in series with the gate, then adjust the gate voltage upward to pass a current through a 10 Ohm resistor in the drain cct that matches your intended application. Via a big Cap (100uF min) directly connected to the gate, feed the gate of the fet (max swing +/_ 5V, or you may kill the fet) a sweep of sine (or square) waves at a level high enough to display a 1V signal across the 10R load resistor at --say 20 HZ---then sweep through to 100KHz...You will see some really nasty stuff. Watch as you approach resonances that you don't damage your test gear or blow the fet away. The fet and the choke will begin to function as a class E amplifier, generating swings in excess of 3- 4 times Vcc.
That is why those nice audio psu chokes cost so much...being a real component, they do have resonances, but they are much better behaved.
A MUCH better alternative is to use a FET and a cap in a cap multiplier- this basic circuit element works like a very well mannered choke and probably takes up a lot less space and costs way less - even with a decent heat sink.
I have tried to upload a schema to help, just relocate the end of the 1 Meg to ground, you may have to play with the resistor values a bit to limit dissipation and get the voltages you want. Remember, this is NOT a series pass regulator, it just cleans up what passes through..
Perhaps someone who has more time could provide links to amplified cap/ cap multiplier/ gyrator-esque devices.
From memory Rod Elliot has a nice primer on his site, but I don't know if this is permissible to mention this here.
Any way, good luck- the time spent getting your head around a cap multiplier will be well spent- you can use a cheap choke infront of Cap multipliers to feed 2 channels and get amazing results for a fraction of the cost of traditional Chokes etc.. I do not mean to be a smart alec, test those flouro ballasts, then build a decent cap multiplier and feed it all sorts of crap and see what gets through..
Regards, Glens
In addition to the concerns raised by the other gentlemen, you should consider this: a decent audio power supply choke will be constructed to minimise the self-capacitance between the windings, its flux leakage and susceptibility. This is significant, because the self-cap is in parallel with the inductance you want to use to reduce the power supply noise etc. This "cap" will do several bad things- it will form a parallel resonant circuit at several frequencies- quite a bit more complicated than the basic text book stuff.
Test this for yourself: connect a good low impedance power supply (12V DC is fine) to a simple Nfet power stage. Play safe - fuses etc-- 1 R in source lead, 10K from gate to source, bias the gate at 0 V with a 100K pot across the rails and a 10 K resistor in series with the gate, then adjust the gate voltage upward to pass a current through a 10 Ohm resistor in the drain cct that matches your intended application. Via a big Cap (100uF min) directly connected to the gate, feed the gate of the fet (max swing +/_ 5V, or you may kill the fet) a sweep of sine (or square) waves at a level high enough to display a 1V signal across the 10R load resistor at --say 20 HZ---then sweep through to 100KHz...You will see some really nasty stuff. Watch as you approach resonances that you don't damage your test gear or blow the fet away. The fet and the choke will begin to function as a class E amplifier, generating swings in excess of 3- 4 times Vcc.
That is why those nice audio psu chokes cost so much...being a real component, they do have resonances, but they are much better behaved.
A MUCH better alternative is to use a FET and a cap in a cap multiplier- this basic circuit element works like a very well mannered choke and probably takes up a lot less space and costs way less - even with a decent heat sink.
I have tried to upload a schema to help, just relocate the end of the 1 Meg to ground, you may have to play with the resistor values a bit to limit dissipation and get the voltages you want. Remember, this is NOT a series pass regulator, it just cleans up what passes through..
Perhaps someone who has more time could provide links to amplified cap/ cap multiplier/ gyrator-esque devices.
From memory Rod Elliot has a nice primer on his site, but I don't know if this is permissible to mention this here.
Any way, good luck- the time spent getting your head around a cap multiplier will be well spent- you can use a cheap choke infront of Cap multipliers to feed 2 channels and get amazing results for a fraction of the cost of traditional Chokes etc.. I do not mean to be a smart alec, test those flouro ballasts, then build a decent cap multiplier and feed it all sorts of crap and see what gets through..
Regards, Glens
Keit, I think if you took the time to test a part before bagging it then this post would better off. They are relatively low inductance compared to 10-20H behemoths, but that doesn't make them a bad choke.
What chokes have interleaved laminations ??? The only ones I know of are swinging type EI where the interleaving is not 1:1 as per a normal transformer aiming at the least effective magnetic air gap.
They test fine as an inductance at 100Hz at relatively high levels of DC current, so to me they are a winner.
I would be thinking their leakage field would actually be less than the same inductance/DC rated choke wound on a typical EI lamination core, as the unenclosed winding section is pretty damn small, and the effective magnetic core air gap appears to be internal to the core, and not external as per an EI format. But I haven't done any comparative testing so haven't really proffered an opinion on that matter til now.
Last edited:
Trobbins, It's amusing when you accuse me of posting without first testing, then in your last para you say you didn't test either.
Perhaps you should take more care reading posts before you bag them.
What chokes have interleaved laminations? Well, certainly not power filter chokes meant to carry DC. But chokes designed for applications where there is no DC almost always DO have interleaved laminations, which led another person to post in a referenced thread above that he would expect flouro ballasts to be unsuitable. My point was that as Australian made fluoro ballasts do not have interleaving anyway (saving cost) and the stackes are welded, they in effect are magnetically gapped anyway, so they might well work better than expected for a choke design for AC being made to carry DC. But not as good as a purpose designed choke.
The leakage magnetic field WILL be pretty bad, actually for two reasons: 1) the core is a long skinny thing, making the magnetic path long. You get less leakage the closer the core is to a round loop. 2) the welding ot the two stacks together, which results in localised deterioration of permeability, as I said.
You may or may not be aware that when cheap Asian imports of transformers and chokes started in about 1969-70, the Australian transformer manufactuer Ferguson had trouble competing and decided to make a line of power transformers on their flouro ballast high volume production line machinery. They were known as the PL-series and were made in 3 sizes: PL20 (20VA), PL40 (40VA), and PL80 (80VA). As I recall, there was also an PL60, rarely encountered, and a few custome sizes available only to other manufacturers in bulk orders.
The PL80's were exactly the same size as their standard twin flouro ballast. The PL40 was a bit more than half the length, and the PL20 a bit more than half again.
All of them were notorious for high leakage flux. The PL20's were bad for this, the PL40's were worse, and the PL80's really bad - as you should expect from the elongated shape.
But we all used them anyway, because they were cheap, readily available, and low profile. Once parts dealers like Dick Smith, Altronics, etc, starting importing Asian transformers made with grain oriented silicon steel, and so pretty small anyway, that was the end of Ferguson and their nasty ballast-built power transformers.
You say ballasts test fine. But you didn't say how you tested them. In circuit or on an inductance tester providing the correct amount of DC? Such testers are rare beasts. Hobbyists don't generally have them. Even professional electronic engineers don't usually have them, and when needed cobble a test circuit up from parts and gear on hand. As I said before, try a resistor in an in-cuircuit test, you might find it works just as well.
I can't see the sense of using flouro ballsts for another reason. Professional enginners stooped using chockes when large microfarad capacitors became available in the early 1960's - they are a lot cheaper than chokes, take up less room, and don't put out heat from I^2.R loss.
But it is reasonable to use a choke if you are restoring an old amp or yoy want to replicate the look. In neither case does a 1980's flouro choke look the part. And a wirewound resistor costs 30 cents.
Last edited:
I referenced the test technique used in the link in post #4.
The 'modern' magnetic flouro choke, to which I believe is the context of this thread, is an extremely optimised choke for its application. I haven't torn down such a choke as yet, but a patent (EP0047266-A1) does show that the magnetic core gap is internal to the core, and that one major design goal of a flouro choke is minimal leakage field, as that can contribute to audible noise from interaction with metal battons.
I can appreciate that people who have had bad experiences with a generic part often keep clear of them in to the future, but I'm pretty sure that the 'modern' flouro chokes that are still being made are probably noticeably different from the old PL types.
LC filtering can have its influence in a guitar amp due to the changed sag dynamics - there is no one optimal design when it comes to guitar amps.
The flouro choke is surprising compact, and I have fitted them under the chassis of old amps, as well as behind the main transformer, so their aesthetical appeal or otherwise can be made a mute point.
The advantage of using a 60 ohm DCR choke over a 1.2kohm resistor to derive B+ in a power amp have two very different outcomes.
The 'modern' magnetic flouro choke, to which I believe is the context of this thread, is an extremely optimised choke for its application. I haven't torn down such a choke as yet, but a patent (EP0047266-A1) does show that the magnetic core gap is internal to the core, and that one major design goal of a flouro choke is minimal leakage field, as that can contribute to audible noise from interaction with metal battons.
I can appreciate that people who have had bad experiences with a generic part often keep clear of them in to the future, but I'm pretty sure that the 'modern' flouro chokes that are still being made are probably noticeably different from the old PL types.
LC filtering can have its influence in a guitar amp due to the changed sag dynamics - there is no one optimal design when it comes to guitar amps.
The flouro choke is surprising compact, and I have fitted them under the chassis of old amps, as well as behind the main transformer, so their aesthetical appeal or otherwise can be made a mute point.
The advantage of using a 60 ohm DCR choke over a 1.2kohm resistor to derive B+ in a power amp have two very different outcomes.
Chokes bring their own benefits which large caps do not have. The filtering of high frequency hash is the main one. The suppression of massive charging spikes when using large caps and the radiated EMF which they can produce is another.
The reality from those who have actually used them is that they work as filters, though admittedly not as well as proper PS designed units. They do not suffer from the problems of radiated hum as suggested. How can I say that - my system will sound objectionable when hum is at 1-2mV of hum at the speaker. I have used these chokes in two preamps (one with a 400V supply) and the output is absolutely flat with no signal so there is no problem of radiated hum and there is definitely at least 1H of inductance remaining after DC effects (this has been measured by multiple forum members). This would be an inadequate amount of inductance for an old fashion small cap power supply , where the choke does all the heavy lifting, but combined with large post choke filter caps they are more than effective. I have even used them in valve rectified power supplies with post choke caps in the order of 400+uf, showing that they are functioning better than an equivalent low DCR value resistor.
At the very least they can be considered as effective cheap power resistors - but the reality is that they perform far better than this.
I have also used them on the top of a case sprayed to the case colour, and they are actually quite aesthetically pleasing if you apply a little though to their location.
You have little to lose by trying this in a circuit.
Shoog
The reality from those who have actually used them is that they work as filters, though admittedly not as well as proper PS designed units. They do not suffer from the problems of radiated hum as suggested. How can I say that - my system will sound objectionable when hum is at 1-2mV of hum at the speaker. I have used these chokes in two preamps (one with a 400V supply) and the output is absolutely flat with no signal so there is no problem of radiated hum and there is definitely at least 1H of inductance remaining after DC effects (this has been measured by multiple forum members). This would be an inadequate amount of inductance for an old fashion small cap power supply , where the choke does all the heavy lifting, but combined with large post choke filter caps they are more than effective. I have even used them in valve rectified power supplies with post choke caps in the order of 400+uf, showing that they are functioning better than an equivalent low DCR value resistor.
At the very least they can be considered as effective cheap power resistors - but the reality is that they perform far better than this.
I have also used them on the top of a case sprayed to the case colour, and they are actually quite aesthetically pleasing if you apply a little though to their location.
You have little to lose by trying this in a circuit.
Shoog
Last edited:
Thanks Shoog, After all the excellent information I got from the other forum members, it's refreshing to have someone who has actually tried these in practice and can offer suggestions as to the value of caps required. I think I saw a section dealing with this in Morgan Jones's book and the inherent resonant frequencies.....