Having a good central ac power source is always a good place to start. I can't tell you guys how many people plug in interconnected equipment into whatever outlets they can find having God knows what ground potentials and wonder why they have problems? Not implying anything lumanauw, but it happens. For instance- at a very well known theatrical sound shop I work for here in NY, much of the gear is self-powered and with a central distro, there are rarely any ground loop problems with the rigs. Every peice of equipment with a ground lug is appropriately plugged in, and all signal interconnects are grounded all the way through. Of course ground loops do happen for often just plain wierd reasons and we always carry iso transformers just in case. But in theory if everything is at the same ground potential and the equipment is behaving, there shouldn't (operative word here) be problems.
In Spain in modern buildings you usually find earthed outlets only in kitchens and bathrooms. In living rooms and nearly all other rooms outlets usually have no earth connection
Most consumer electronics come without earth connection, examples : TV sets, CD/DVD/VCR players, compact stereos, audio amplifiers and most hi-fi equipment
Nearly all small home appliances come without earth connection [only big ones like washing machines, fridges, dishwashers or microwave ovens provide it]
Some PA equipment comes without earth connection, examples : Mixing consoles, equalizers, reverb modules, compressors/expanders, small amplifiers, etc...
In power amplifiers, big toroidal transformers are the part to blame since they are the main cause for ground loops and 'humm' due to their excessive capacitance from primaries to secondaries. They usually show too high capacitive leakage even if they pass the 3750V isolation test. [This is the reason why I hate toroidal mains transformers, SMPS tend to have much lower leakages and capacitances]
Most consumer electronics come without earth connection, examples : TV sets, CD/DVD/VCR players, compact stereos, audio amplifiers and most hi-fi equipment
Nearly all small home appliances come without earth connection [only big ones like washing machines, fridges, dishwashers or microwave ovens provide it]
Some PA equipment comes without earth connection, examples : Mixing consoles, equalizers, reverb modules, compressors/expanders, small amplifiers, etc...
In power amplifiers, big toroidal transformers are the part to blame since they are the main cause for ground loops and 'humm' due to their excessive capacitance from primaries to secondaries. They usually show too high capacitive leakage even if they pass the 3750V isolation test. [This is the reason why I hate toroidal mains transformers, SMPS tend to have much lower leakages and capacitances]
Yeah, chances are that's what's causing at least some of the problem. Not enough current on one circuit to power the whole rig? It really would be worth your while to run power fom a central source via extention cords if possible. If you must draw from several different outlets, some good quality iso transformers should do the trick. And if your runs are over 10 feet or so from the board to the speakers, you really should be sending a balanced signal if the speakers have a balanced input. Unbalaced signals can become unreliable and noisy after a short distance.
yeah...they're called engine noise removers or something like that...though I wouldn't really understand why a running rngine and alternator can cause such problems?? I'm not really sure if they are transformers though...an example here...
An externally hosted image should be here but it was not working when we last tested it.
Hi, Joan,
You are right. I WANTED to DIY line transformer. The cheap line transformer for car audio is exactly what I'm using right now, with bad sound. The highs are missing.
Looking at the net, Jensens are too expensive, so I wanted to DIY. Could you help me with the formulas? I can buy any size magnet wire and any size power transformer core here.
Someone said something about capacitance in line transformer. Is this the difficult thing to handle if we go DIY? Needs certain core material or wire material?
You are right. I WANTED to DIY line transformer. The cheap line transformer for car audio is exactly what I'm using right now, with bad sound. The highs are missing.
Looking at the net, Jensens are too expensive, so I wanted to DIY. Could you help me with the formulas? I can buy any size magnet wire and any size power transformer core here.
Someone said something about capacitance in line transformer. Is this the difficult thing to handle if we go DIY? Needs certain core material or wire material?
lumanauw said:
You had bad luck, try another one, they are cheap.
My brother has one for car audio from the DVD player to the amp, because of problems (noise) with cable TV.
Believe me, no difference at all with or without it, and the noise is gone.
It may be pure luck, a cheap one that's good...
and making a trafo like that ain't exactly easy either...maybe you would want to try a ground loop breaker as described in this page...
http://sound.westhost.com/earthing.htm
http://sound.westhost.com/earthing.htm
Back to topic
Hi lumanauw!
If you have access to ferrites, then you do not need to worry about the
resonance frequency of the ferrite material itself. Their resonance frequencies usually are high enough. Drawback is that their saturation flux density is typically somewhere between 300mT and 400mT...
The thing which may cause issues is the distributed winding capacitance withing each winding itself. This can be kept low if you
keep the number of turns as low as possible.
If we look at the formula for the required number of turns:
N = U / (4.44 x f x B x Ae)
Then we see that we should look for a core shape, which provides large cross section area. This pushes towards RM-shapes, EP-shapes, PM-shapes...., less fortunate are traditional E-shapes...
Furtheron I am afraid of the capacitance within one winding.
Less critical should be the coupling capacity between the two windings (I think.... please everybody correct if wrong).
Also isolation requirements are low.
So I would try to wind secondary and primary with paralleled wires in order to achieve a good coupling. Make sure that you connect primary and secondary in the same way to your connectors.
I.e. start of winding to signal and end of winding to ground.
Other wise you will invers the signal and (much worse) if not connected in same way then you should also be afraid of the coupling capacitance between primary and secondary...!
Which cores do you have on hand or can get?
Which driving impedance do you have? Which levels?
Probably normal line signal?
So the inductive impedance of the XFR should not drop below
10kOhms even at the lowest desired signal frequency.
Also we should be able to handle at least 1Vrms at 40Hz without saturation...., right?
Try to find out and tell wich cores you have....
May be I can go for a rough design proposal and explain my thoughts.
Starting with this, then some more trial and error + good luck
may bring success.
Good luck
Markus
Hi lumanauw!
If you have access to ferrites, then you do not need to worry about the
resonance frequency of the ferrite material itself. Their resonance frequencies usually are high enough. Drawback is that their saturation flux density is typically somewhere between 300mT and 400mT...
The thing which may cause issues is the distributed winding capacitance withing each winding itself. This can be kept low if you
keep the number of turns as low as possible.
If we look at the formula for the required number of turns:
N = U / (4.44 x f x B x Ae)
Then we see that we should look for a core shape, which provides large cross section area. This pushes towards RM-shapes, EP-shapes, PM-shapes...., less fortunate are traditional E-shapes...
Furtheron I am afraid of the capacitance within one winding.
Less critical should be the coupling capacity between the two windings (I think.... please everybody correct if wrong).
Also isolation requirements are low.
So I would try to wind secondary and primary with paralleled wires in order to achieve a good coupling. Make sure that you connect primary and secondary in the same way to your connectors.
I.e. start of winding to signal and end of winding to ground.
Other wise you will invers the signal and (much worse) if not connected in same way then you should also be afraid of the coupling capacitance between primary and secondary...!
Which cores do you have on hand or can get?
Which driving impedance do you have? Which levels?
Probably normal line signal?
So the inductive impedance of the XFR should not drop below
10kOhms even at the lowest desired signal frequency.
Also we should be able to handle at least 1Vrms at 40Hz without saturation...., right?
Try to find out and tell wich cores you have....
May be I can go for a rough design proposal and explain my thoughts.
Starting with this, then some more trial and error + good luck
may bring success.
Good luck
Markus
It is not DIY but here are some other suppliers with a range of prices.
http://www.diyaudio.com/forums/showthread.php?threadid=35347
http://www.diyaudio.com/forums/showthread.php?threadid=35347
Hi, all,
Thanks for the help. Tomorrow I will go to local electronic market here and find out what ferrites I can buy here. I have used EE19 (4.5x4.75mm), EE20(5.6x5.6mm), EE25(6.25x6.25mm). Also I can buy EE core without label but the core size is(4.95x9.95mm)
Those are small ferrites. Big Ae ferrite I can buy those used for flyback for TV. I also can buy toroidal ferrites in various range of sizes, but I cannot wind so much winding in toroids. Maybe we'll stick to EE or EI cores.
Any idea what Ae will be needed for line transformer? Will it about the size of EE20 or will size about TV flyback transformer?
First question. Will ferrites better than metal core? What is the one in Jensen core, and also what is the one inside the cheap car audio isolator? Tube amps usually use special metal instead of ferrites?
Another question. What will the magnetwire size? 0.1mm is enough? Can use ordinary magnetwire or needs special properties?
Thanks for the help. Tomorrow I will go to local electronic market here and find out what ferrites I can buy here. I have used EE19 (4.5x4.75mm), EE20(5.6x5.6mm), EE25(6.25x6.25mm). Also I can buy EE core without label but the core size is(4.95x9.95mm)
Those are small ferrites. Big Ae ferrite I can buy those used for flyback for TV. I also can buy toroidal ferrites in various range of sizes, but I cannot wind so much winding in toroids. Maybe we'll stick to EE or EI cores.
Any idea what Ae will be needed for line transformer? Will it about the size of EE20 or will size about TV flyback transformer?
First question. Will ferrites better than metal core? What is the one in Jensen core, and also what is the one inside the cheap car audio isolator? Tube amps usually use special metal instead of ferrites?
Is this about special winding techniques?
OK, understand that.
Yes, normal signal. Like the one comes out from CD player, or DVD player. But I wanted it universal, also for the ones that comes out from preamps, voltage maybe higher than CD player.
OK, understand that.
Is this standard requirement? I will follow standard.
Another question. What will the magnetwire size? 0.1mm is enough? Can use ordinary magnetwire or needs special properties?
Just calculating a little.
The Faraday law for determining N for certain core size.
N=Vx1x10(exp8)/(4.44xAexBxf).
Chocoholic, there is 1x10exp(8) in the equation, right? Ae in cm2, B in Gauss, f in Hz.
If the ferrite size is 8x8mm, with B=3000gauss, f=20hz, V=1V, then the winding number N=587turns.
But I dont see any Kohm here. Usually line transformer spec like Jensen's is 10k:10k.
The equation for knowing impedance from L is R=wL=2xpixfxL
To get 10K in 20hz the L must be 79Henry?
To get 1H it is very big windings. We must get 79H?
Or the specs in the Jensens are 10k:10k in what Hz? Is it in 20hz?
The Faraday law for determining N for certain core size.
N=Vx1x10(exp8)/(4.44xAexBxf).
Chocoholic, there is 1x10exp(8) in the equation, right? Ae in cm2, B in Gauss, f in Hz.
If the ferrite size is 8x8mm, with B=3000gauss, f=20hz, V=1V, then the winding number N=587turns.
But I dont see any Kohm here. Usually line transformer spec like Jensen's is 10k:10k.
The equation for knowing impedance from L is R=wL=2xpixfxL
To get 10K in 20hz the L must be 79Henry?
To get 1H it is very big windings. We must get 79H?
Or the specs in the Jensens are 10k:10k in what Hz? Is it in 20hz?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.