P2P Redd 47 - a few questions

[chrispsound]

Sorry about the schematic, it is on page 7 of this thread  posted by earthsled.  For the purposes of testing I am just using  the power supply(schematic is on my original post) with just 2 capacitors a 1k 5w resistor and a 470k bleeder and I am not using zeners.  Since my toroid PT has many taps I am now using the 220vAC tap.  My power supply is now at 300vDC unloaded after rectification and filtering.  With the E88CC pulled I am getting about 297v at pin 1 and 6 and no voltage on any other pin at tube socket except heater ac at pin 4 and 5. I am getiing 299 volts at the top of the plate resistor. I think my tube may be bad?  Thanks for the help so far.

The E88CC was bad, drawing too much current.  Put a new tested tube in and it is now functioning correctly.  Nothing like a good problem to force learning through a thick skull.  I guess I am one of those yahoo's that doesn't fully understand ohms law and conceptual electronics that is heavily armed with a will and a soldering iron(trying to drop current, duhhh!).  I am one step closer to enlightenment. Thanks Again.  ChrisP

 

[earthsled]

You do need a 100mA version.

I'm assuming this would apply to 2CH of REDD.47 using the bridge config, rather than full wave CT. -- is this correct?

The same power formula in both cases. So, for the bridge, the voltage is 380/1.4 and the current is 60mx1.4.For the full-wave CT, the voltage is 2x380/1.4 and the current is 1/2.60mx1.4. The actual mean current used to evaluate resistive losses is slightly higher.

Here are my calculations for HT transformer requirements:

Bridge
1CH of REDD.47 = 269V @ 42mA (51mA w/ safety margin)
2CH of REDD.47 = 269V @ 85mA (102mA w/ safety margin)

Full Wave CT
1CH of REDD.47 = 537V @ 21mA (25mA w/ safety margin)
2CH of REDD.47 = 537V @ 42mA (51mA w/ safety margin)

Please let me know if there are errors here.

Thanks!

 

[abbey road d enfer]

You must increase all currents by 1.4
20mA is the DC current of 1 channel, the AC current must be 28mA.

 

[earthsled]

20mA is the DC current of 1 channel, the AC current must be 28mA.

The schematic says 30mA DC per channel, so AC current must be 42mA per channel -- right?

???

 

[abbey road d enfer]

Yes, you're right. Your figures are correct. I read too hastily.

 

[earthsled]

Okay, great. 

For what it's worth, the original REDD.47 P2P layout makes use of the full wave w/ center tap config. The suggested Allied 227-0081 transformer seems highly over-rated at 120mA, considering only 25mA per channel is necessary. This transformer could easily power 4 channels!

 

[abbey road d enfer]

Okay, great. 

For what it's worth, the original REDD.47 P2P layout makes use of the full wave w/ center tap config. The suggested Allied 227-0081 transformer seems highly over-rated at 120mA, considering only 25mA per channel is necessary. This transformer could easily power 4 channels!

It is not uncommon to use an overspec'd component when it's already in the system or when not certain of the final outcome; maybe they were thinking of using one PSU for 2 or 3 channels.
In my current company, we use only two models of mains xfmrs, rated 15 and 30VA. We have some units that would accomodate an 8VA model. The price and size difference are so slight we never bothered. Same for elcaps, we use only 100uF and 470uF. Sometimes when 1uF would be enough, we put a 100uF. No difference in price, less possible mistakes.

 

[earthsled]

It is not uncommon to use an overspec'd component when it's already in the system or when not certain of the final outcome; maybe they were thinking of using one PSU for 2 or 3 channels.
In my current company, we use only two models of mains xfmrs, rated 15 and 30VA. We have some units that would accomodate an 8VA model. The price and size difference are so slight we never bothered. Same for elcaps, we use only 100uF and 470uF. Sometimes when 1uF would be enough, we put a 100uF. No difference in price, less possible mistakes.

Good points. Especially considering the Allied transformer is very affordable and has the advantage of being "off-the-shelf".

My plan has changed from using an off-the-shelf guitar amp power transformer to ordering a custom-wound model form Edcor. This being my first custom power transformer, I'm very concerned about getting the AC currents correct. Of course, a transformer that is under-rated is far worse than one that is over-rated. Although, an over-rated transformer can cause issues too.

Sowter and Hammond seem to share the same guidelines for specifying AC secondary current (see attached). The notes on this document suggest a "safety margin" has been added to the equations for capacitor input load PSUs. "Included in the formulas higher peak to peak capacitor charging current in the filter."

Abbey's suggestions for AC secondary current:
Full Wave: Iac = Idc x 0.7 (Iac = Idc x 0.8 w/ safety margin)
Bridge: Iac = Idc x 1.4 (Iac = Idc x 1.7 w/ safety margin)

Hammond / Sowter suggestions for AC secondary current:
Full Wave: Iac = Idc x 1
Bridge: Iac = Idc x 1.61

With the safety margin taken into account, the guidelines appear fairly close.

 

[earthsled]

Here's another document on the subject. In this case, the conclusion allows a bit more than the Hammond / Sowter guidelines:

Full Wave: Iac= Idc x 1 to 1.2
Bridge: Iac = Idc x 1.6 to 1.8

Cheers!

 

[earthsled]

The folks at Edcor tell me their power transformers are designed with flux of 1.4 to 1.5 Tesla. I have no idea if this is considered low or high compared to other PTs. Has anyone had experience with this?

Thanks!

 

[ricardo]

The folks at Edcor tell me their power transformers are designed with flux of 1.4 to 1.5 Tesla. I have no idea if this is considered low or high compared to other PTs. Has anyone had experience with this?

Mild steel will show signs of saturation at 1.2T ie if the peak flux reaches that level, the magnetising current will rise sharply as you raise the input voltage.

A PT designed to the limit for 60Hz will be stressed at 50Hz as the flux is 20% higher for the same voltage.

All this is dependent on the grade of steel used.  You need to look up an old book on transformer design and find out what Edcor use.  Old Man Sowter did a very good article in JAES on this and I think the Radio Designers Manual has a good treatment.

 

[earthsled]

Thanks for the info Ricardo -- I find it very interesting that 50Hz will cause more flux compared to 60Hz. Edcor's website claims they use at least 92% M6 line grain oriented steel cores. But, I'd mainly like to know how the flux specs from Edcor compare to a typical off-the-shelf model. Abbey mentioned earlier that some manufactures will wind custom transformers to achieve a lower stray flux for sensitive audio gear. Is 1.4 to 1.5 Tesla fairly normal?

Thanks!

 

[abbey road d enfer]

Thanks for the info Ricardo -- I find it very interesting that 50Hz will cause more flux compared to 60Hz. Edcor's website claims they use at least 92% M6 line grain oriented steel cores. But, I'd mainly like to know how the flux specs from Edcor compare to a typical off-the-shelf model. Abbey mentioned earlier that some manufactures will wind custom transformers to achieve a lower stray flux for sensitive audio gear. Is 1.4 to 1.5 Tesla fairly normal?

Thanks!

It seems that the accepted value for EI transformer is 1T. Considering the better magnetic continuity and better flux concentration, 1.5T seems to be right. Clearly, toroids operate at higher induction, closer to saturation, which makes them less tolerant to overvoltage. For the same weight of iron, toroids can pass 1.5 times the power. You may ask a custom toroid to operate at 1T, that means you'll have a 30W core for a 20W xfmr.

 

[ricardo]

Is 1.4 to 1.5 Tesla fairly normal?

I don't know cos I'm not a Transformer Guru and it depends on the material.  In my field of expertise, speakers, mild steel is commonly run at a max of 1.2T but for special applications, I've used special "Swedish Steel" at 1.6T

I've no idea if 1.5T is OK for "92% M6 line grain oriented steel".  The info will be in Sowter & RDH

If this is important to you, look at the Magnetisation Current (AC current with no load) at various input mains voltages.  A good UK maker will specify Imag at 230V and also at the max 253V.  Look at the specs of some reputable manufacturers for the same VA to get an idea of what is "normal" and what is iffy.

Imag is a good indication of saturation stray flux but only for the same type of construction; ie compare toroids with toroids.

Most modern EU gear use 2x115V primaries.  When used in the US, the voltage is higher but 50Hz design easily compensates for US 120V 60Hz.  The converse is not always true.

Some countries don't control their mains accurately and transformers might saturate & blow fuses.  In my previous life, the worse I've found is Malaysia.  You can't afford iffy transformers if you want gear to operate there.

 

[earthsled]

It seems that the accepted value for EI transformer is 1T. Considering the better magnetic continuity and better flux concentration, 1.5T seems to be right. Clearly, toroids operate at higher induction, closer to saturation, which makes them less tolerant to overvoltage. For the same weight of iron, toroids can pass 1.5 times the power. You may ask a custom toroid to operate at 1T, that means you'll have a 30W core for a 20W xfmr.

I probably should have mentioned that Edcor makes EI transformers rather than toroids. If 1T is the accepted value, then 1.5T would be a bit high for an EI type, right? I'm assuming a higher number means more stray flux and potentially increased hum in a mic pre. Practically speaking, would it be a bad idea to have the PT onboard the preamp chassis with a flux rating of 1.5T?

Thanks!

 

[abbey road d enfer]

It seems that the accepted value for EI transformer is 1T. Considering the better magnetic continuity and better flux concentration, 1.5T seems to be right. Clearly, toroids operate at higher induction, closer to saturation, which makes them less tolerant to overvoltage. For the same weight of iron, toroids can pass 1.5 times the power. You may ask a custom toroid to operate at 1T, that means you'll have a 30W core for a 20W xfmr.

I probably should have mentioned that Edcor makes EI transformers rather than toroids. If 1T is the accepted value, then 1.5T would be a bit high for an EI type, right? I'm assuming a higher number means more stray flux and potentially increased hum in a mic pre. Practically speaking, would it be a bad idea to have the PT onboard the preamp chassis with a flux rating of 1.5T?

Thanks!

I guess you understand I can't give you a definite binary answer, but I would say that it is subject to experimentation. Coupling between transformers can be tuned-out by distance, orientation and shielding. Putting the xfmr as far as can be from the sensitive stages, allowing rotation of the xfmr and the addition of a judiciously placed mild-steel screen (or mu-metal) may be the answer. Or not.
Just like saying you can get 500bhp from a small block. It is feasible, does not mean everybody can.
It would be easier to have a low-induction xfmr. But still you would have to experiment with its placement.

 

[earthsled]

It would be easier to have a low-induction xfmr. But still you would have to experiment with its placement.

Thanks Abbey. Points taken. I've asked Edcor if they have the capability to build the transformer with a lower flux rating. At one point you mentioned that Leo Fender had transformers built with reduced flux for his amps. Any idea what kinds of flux ratings were specified? I'm thinking that if 1T is normal, then 0.8 or lower could be considered "reduced flux" for a typical EI power transformer in the 50 to 100VA range -- would you agree?

I appreciate your help!

 

[CJ]

<it seems that the accepted value for EI transformer is 1T...>

1.2 T for magnesil, or 29gaM6, 4%silicon steel, but we ain't picky, 

"operating flux swing" will obviously be lower than the saturation flux level for all the different materials.

<considering the better magnetic continuity and better flux concentration, 1.5T seems to be right...>

actually,  because of this better magnetic continuity and better flux concentration, you have to be more careful with the flux level, as the toroid core will tend to reach saturation easier, think of a square BH curve, once you hit the top, it folds over into a straight line.
on the other hand, EI cores have a natural air gap built in compared to a tape wound core.
every time you do a turn with a piece of tape, you are adding a lam, only without the butt joint, so with each turn, an air gap is eliminated.
Also, you can only clamp an EI core so hard, there will always be a little space in between lams, which also adds to the air gap, which makes the BH curve of an EI core less square for the same material wound as a toroid.
this is why toroids do not tolerate DC as well as an EI core.
you can really get all the domains lined up with a toroid as the air gap is 10^-6 cm for a lot of these cores. this is why your variac trips the breaker sometimes, you turn it off at the peak of the sine wave and all the domains are aligned to saturation, you flip the power back on and the saturated core allows massive inrush current, just stick a light bulb in series like the dead people did.

"clearly, toroids operate at higher induction, closer to saturation, which makes them less tolerant to over-voltage...>

see previous paragraph.

<for the same weight of iron, a toroid can pass 1.5 times the power..."

not because of the of the increased flux density,

more due to area product vs current density for the various cores, Kj and all that.
not PCP, i quit that stuff years ago. 

 

[shabtek]

good 'ol CJ

 

[CJ]

here is some toroid stuff,

me bad, looks like the flux density is higher, i'm such a loser, ;D

The Toroidal Core

At the heart of the toroidal is a highly efficient donut shaped core. To construct the core, grain-oriented silicon-iron is slit to form a ribbon of steel which is then wound, like a very tight clock spring. The result is a core in which all of the molecules are aligned with the direction of flux. Molecules not aligned with the flux direction increase a core's reluctance (the capacity for opposing magnetic induction), degrading performance to the level of common steel when the molecules are 90 degrees out of phase. EI laminated cores, which are stamped from grain-oriented Si-Fe, may have as much as 40% of the total core area perpendicular to the ideal grain direction, with another 40% acting only as a return flux path. This more efficient use of the core material in a toroidal can result in a size and weight reduction of up to 50% (depending on power rating), allowing the design engineer to innovate by exploiting the toroidal's small size, low weight, ease of mounting, and flexible dimensions.

Efficiency

Since toroidal cores are constructed of a continuously wound ribbon, there is virtually no air gap. The windings are evenly wrapped over the entire core allowing the transformer to operate at a higher flux density than in standard transformers. Toroidal transformers can operate at 1.6 to 1.8 Tesla (16,000 to 18,000 Gauss) while EI cores are limited to 1.2 to 1.4 Tesla (12,000 to 14,000 Gauss). The magnetic flux of the windings is oriented in the same direction as the grain-oriented core, thus achieving very high electrical efficiencies. Efficiency is a measure of a transformer's ability to deliver the input power to the load. Efficiency is expressed as a percent by:

% = ( PO / PI ) x 100

where; PO = Output power, PI = Input power, % = Efficiency

Also, standby losses are greatly reduced under no-load operation due to the lower magnetizing currents required by the toroidal core.

Stray Magnetic Fields

The primary cause of leakage flux from any transformer is the air gap. Ideally, a magnetic circuit should have no air gap. In traditional transformers with EI laminations stacked to form the core, the air gap at the junction of the I and the E is the source of most of the leakage flux. This flux strays into the surroundings due to the high reluctance of the air and the concentration of flux in the laminations. For the same reasons, mounting holes and grooves in the laminations also cause a small amount of leakage flux. The tape wound cut-C core is an improvement; but there is still an large air gap causing unwanted stray flux. Since toroidal cores are wound from a continuous ribbon of steel, stray fields from air gaps are eliminated.

In addition, the windings of the toroidal transformer uniformly encase the core in copper. This results in a natural magnetic screening effect which, in combination with the elimination of the air gap, results in an 8:1 reduction of radiated magnetic field over an equivalent rated EI transformer. The windings covering the solid ring core also help reduce magnetostriction -- the main source of acoustic "hum" in standard transformers. Audible noise can be reduced even further by varnish impregnating the toroidal core and/or the copper windings.

A toroidal transformer has its winding spread over the entire periphery of the core. This results in largcr area for dissipating the heat. Till recently all such small low voltage type transformers needs were met almost universally by traditional or conventional El transformers. Now materials are easily available, so it has been possible to develop 'LOW VOLTAGE TOROIDAL TRANSFORMER' giving a whole set of advantages and wide range of applications in place of conventional transformers.