Were gonna see if it's wearin the colors.
St. Ives vs Marinair-Line In
- Thread starter CJ
- Start date
Help Support GroupDIY Audio Forum:
Were gonna see if it's wearin the colors.
I have the Marinair 31267 tore up already, so no harm done.
That's what I am going to compare this with.
cj
That's what I am going to compare this with.
cj
ah, do tell.
I think they sound different with the small batch I have here. It would be interesting to see a proper technical comparison just to see if there is any good reason why the one would actually sound different from the other.
people have accused me of being insane also, so just bear that in mind.
dave
I think they sound different with the small batch I have here. It would be interesting to see a proper technical comparison just to see if there is any good reason why the one would actually sound different from the other.
people have accused me of being insane also, so just bear that in mind.
dave
DCR a bit different already.
I can't comppletely hack this one, it's gotta go back.
But here are a few pics with the cover off.
Same structure/lams/brass as the green Marinair:
http://vacuumbrain.com/The_Lab/TA/St_Ives/ives_2.jpg
White RTV between lids and potting compound:
http://vacuumbrain.com/The_Lab/TA/St_Ives/ives_3.jpg
More modern G-10 header:
http://vacuumbrain.com/The_Lab/TA/St_Ives/ives_4.jpg
Here are the leakage capacitance figures.
Early black Marinair T-1452 vs the St. Ives:
(trying a new linking thing instead of hassling with a screen capture of an Excel spreadsheet)
No earth shattering differences here.
http://vacuumbrain.com/The_Lab/TA/St_Ives/t-1452_leak_cap_st.ives.htm
But here are a few pics with the cover off.
Same structure/lams/brass as the green Marinair:
http://vacuumbrain.com/The_Lab/TA/St_Ives/ives_2.jpg
White RTV between lids and potting compound:
http://vacuumbrain.com/The_Lab/TA/St_Ives/ives_3.jpg
More modern G-10 header:
http://vacuumbrain.com/The_Lab/TA/St_Ives/ives_4.jpg
Here are the leakage capacitance figures.
Early black Marinair T-1452 vs the St. Ives:
(trying a new linking thing instead of hassling with a screen capture of an Excel spreadsheet)
No earth shattering differences here.
http://vacuumbrain.com/The_Lab/TA/St_Ives/t-1452_leak_cap_st.ives.htm
Leakage Inductance--
Starting to see some differences here. The fact that the leakage capacitance is the same, but the leakage inductance is different in some places makes me wonder if St. Ives for-went the flipping of the mandrels, which is a pain in the rear, but must be done if reverse winding is to be accomplished.
Interesting to note that I could get that 10 mh reading on the Marinair, but the display just blinks on the St. Ives, which means the nmachine is having problems getting a reading.
Here is a Excel page saved as another web page.
Same deal, St. Ives in Red, Marinair in Black:
http://vacuumbrain.com/The_Lab/TA/St_Ives/t-1452_leak_ind_vs_stIves.htm
OK, tomorrow were gonna get down to the real nitty gritty, freq plots, phase shift, B-H curves, saturation levels, mag current waveforms, inductance vs frequency, our latest test:current draw vs frequency, ringing test and a Bode plot of Granny's moustache! :razz:
Later...
cj
Starting to see some differences here. The fact that the leakage capacitance is the same, but the leakage inductance is different in some places makes me wonder if St. Ives for-went the flipping of the mandrels, which is a pain in the rear, but must be done if reverse winding is to be accomplished.
Interesting to note that I could get that 10 mh reading on the Marinair, but the display just blinks on the St. Ives, which means the nmachine is having problems getting a reading.
Here is a Excel page saved as another web page.
Same deal, St. Ives in Red, Marinair in Black:
http://vacuumbrain.com/The_Lab/TA/St_Ives/t-1452_leak_ind_vs_stIves.htm
OK, tomorrow were gonna get down to the real nitty gritty, freq plots, phase shift, B-H curves, saturation levels, mag current waveforms, inductance vs frequency, our latest test:current draw vs frequency, ringing test and a Bode plot of Granny's moustache! :razz:
Later...
cj
This is a side by side shot of a square wave test. Marinair black T-1452 vs St. Ives 31267.
There are some subtle differences. First is 10 kcps sq wave w. 0.002 cap.
St Ives on left.
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar_ives_10khz_cap.jpg
Next, the same test at 100kcps:
St Ives on left.
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar-ives_100khz_cap.jpg
There was a differwent second self resomnance also. 1 volt St Ives next to 357 mv Marinair:
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar-ives_sweep.jpg
Soundguy, your buddies owe you one. Ears of a hound dog I tell ya.
There are some subtle differences. First is 10 kcps sq wave w. 0.002 cap.
St Ives on left.
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar_ives_10khz_cap.jpg
Next, the same test at 100kcps:
St Ives on left.
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar-ives_100khz_cap.jpg
There was a differwent second self resomnance also. 1 volt St Ives next to 357 mv Marinair:
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar-ives_sweep.jpg
Soundguy, your buddies owe you one. Ears of a hound dog I tell ya.
OK, back with more Marinair - St Ives Line in x-former comparisons.
This time it's inductance vs frequency at voltages levels of
200 mv
300 mv
400 mv
500 mv
1 volt and
2 volts rms input to the primaries, full pri and sec used in series.
Henries, in the hundreads, on the left, frequency on the bottom.
You can see why different inductance bridges will give you different numbers, depending on what frequency they take the measurement at.
The core drops out at pretty much at anything over 1 k cps.
The lams have to be different. I am guessing St. Ives used 80 nickel and the black Marinair 50/50.
Here are some graphs with the Marinair highlighted in red, and St. Ives in yellow.
At the 400 mv level and 4 hz, the St Ives is in saturation already, while the Marinair is doing fine. Then as the St Ives comes out of saturation, it reaches max Henries at 10 hz and overtakes the Marinair due to a higher nickel core. You can see the difference in inductance between the two at 10 hz.
The higher perm 80 Ni saturates easier, but once it is out of saturation, it has higher inductance. The 50/50 does not saturate as easy, but does not have as many Henries as the 80 Ni during normal bussiness hours:
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_3.jpg
Here is a chart showing the iron at 1 volt rms input. At 4 hz, the St Ives is really hammered, where as the Marinair is soaked for a bit until 6 hz is reached:
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_5.jpg
So what does all this mean?
The transformers will probably sound different.
One will probably sound "poofier" than the other. :razz:
Here is an Excel file with all the numbers if you want, as well as the rest of the charts.
Note: There are six sheets in that excel file, click on them at the bottom.
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar-ives_inductance.xls
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_1.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_2.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_4.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_6.jpg
This time it's inductance vs frequency at voltages levels of
200 mv
300 mv
400 mv
500 mv
1 volt and
2 volts rms input to the primaries, full pri and sec used in series.
Henries, in the hundreads, on the left, frequency on the bottom.
You can see why different inductance bridges will give you different numbers, depending on what frequency they take the measurement at.
The core drops out at pretty much at anything over 1 k cps.
The lams have to be different. I am guessing St. Ives used 80 nickel and the black Marinair 50/50.
Here are some graphs with the Marinair highlighted in red, and St. Ives in yellow.
At the 400 mv level and 4 hz, the St Ives is in saturation already, while the Marinair is doing fine. Then as the St Ives comes out of saturation, it reaches max Henries at 10 hz and overtakes the Marinair due to a higher nickel core. You can see the difference in inductance between the two at 10 hz.
The higher perm 80 Ni saturates easier, but once it is out of saturation, it has higher inductance. The 50/50 does not saturate as easy, but does not have as many Henries as the 80 Ni during normal bussiness hours:
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_3.jpg
Here is a chart showing the iron at 1 volt rms input. At 4 hz, the St Ives is really hammered, where as the Marinair is soaked for a bit until 6 hz is reached:
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_5.jpg
So what does all this mean?
The transformers will probably sound different.
One will probably sound "poofier" than the other. :razz:
Here is an Excel file with all the numbers if you want, as well as the rest of the charts.
Note: There are six sheets in that excel file, click on them at the bottom.
http://vacuumbrain.com/The_Lab/TA/St_Ives/mar-ives_inductance.xls
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_1.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_2.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_4.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/marives_inductance_6.jpg
Family Hoof
Well-known member
Good stuff, CJ! Verifies what my ears tell me.
Thanks.
Closing in on the completion of the testing for the St. Ives.
Here is a little page on the primary saturation voltage levels between the St. Ives and Marinair. I would have thought that the St. Ives would have a lower saturation voltage since it looked like 80% Ni compared to the MArinair 50/50, but it did not turn out that way, the St Ives had a slightly higher saturation voltage. So I have to think about this one for a while. It's either a lab mistake or something differnt than what my reasoning would indicate. The lams are definately different, as you will see with the B-H curves that are coming right up. Anyway, here is the mag current page:
http://vacuumbrain.com/The_Lab/TA/St_Ives/marinair_ives_saturation.html
Closing in on the completion of the testing for the St. Ives.
Here is a little page on the primary saturation voltage levels between the St. Ives and Marinair. I would have thought that the St. Ives would have a lower saturation voltage since it looked like 80% Ni compared to the MArinair 50/50, but it did not turn out that way, the St Ives had a slightly higher saturation voltage. So I have to think about this one for a while. It's either a lab mistake or something differnt than what my reasoning would indicate. The lams are definately different, as you will see with the B-H curves that are coming right up. Anyway, here is the mag current page:
http://vacuumbrain.com/The_Lab/TA/St_Ives/marinair_ives_saturation.html
Marinair vs St Ives- BH Curve Testing-
There is a big difference in the B-H curves between these two Line In transformers. I do not know how this affects the sound as I have not done a listening test on these two, but I would think you would notice something different on the bottom end.
The B-H curve is a combination between a magnetization curve and an integrated curve. They can be taken with one on the primary and one on the secondary, or both on the primary or secondary. Either way tends to give the same results, as you can saturate the core from either end of the transformer. The amount of steel dosen't change. More turns will require less current to saturate, and vice versa.
I usually use a test setup like this:
It's nice to have the scope grounds isolated by the transformer. Here is a picture of both waveforms before they are combined with the X-Y display function on the scope. The sawtooth is the integrated wave which is taken across the cap, and the other wave is the mag wave, which you might recognize from the previous post. It is taken off the resitor in series with the primary. You must adjust both amplitudes to be the same in order to get a correct BH curve. This is made easier if the scope has fine adjustments on the amplitudes:
Here is what the waves look like when they are assigned X and Y inputs on a non time domained sweep. The first rising edge is the transformer getting energized by the signal. When the core can not handle any more magnetic flux caused by an increasing voltage and thus current through the primary windings, the voltage on the other side of the transformer will no longer rise along with the primary voltage. No matter how much juice you put into the primary, the secondary voltages stays about the same because the magnetic flux in the core is at it's maximum. The rest of the energy is dissapated as a magnetic field out into free space, away from the core. The flat spot on top of the BH curve represents this maxed out voltage.
When the voltage through the primary decreases, the voltage on the secondary decreases also. But not right away. There is a slight delay, which is seen as the space between the first and second vertical lines. This is because the core is still magnetized in a North to South direction, and now you want to magnetize it from South to North. But it has a little memory which must be overcome. Reversing the polarity of the voltage does this, but with a slight delay. The energy lost in switching the magnetic flux around in an opposite direction is called core loss. This is one of the things that chews up your signal in a transformer, especially at the low end.
As the frequncy decreases, this gives the core a longer time to get magnetized. Thus, the lower the frequency, the more core loss, and therefore, the wider the area between the two vertical lines.
Here is the St Ives at 25 hz, and then 15 hz showing the increase in core loss. You have to click on these, don't want to load up the post to much:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/ives_25.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/ives_15.jpg
So it is desirable to have core material that can have it's magnetic flux switched back and forth without using up a lot of energy. You can see this loss on the following BH curve. Remember that the delay in getting the core flux reversed is what seperates the two vertical lines. So the more distance between these lines, the more core loss. In fact, you can calculate the core loss by measuring the area inside the BH curve:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/bh_demo.jpg
OK, so now that your totally confused, lets look at some Marinair vs St Ives Neve Line In transformer curves.
Here is a side by side shot of the Marinair on the left, St Ives on the right, taken with a 20 Hz sine wave at 7.24 volts on each. Don't ask me what that weird lookin curve is from on the St. Ives, as I don't know:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/marives_20hz_sine_7.jpg
Here is a side by side of a 20 Hz square wave, at 7.28 VAC rms input, which is slammin line input iron pretty hard. Not a huge difference here:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/marives_20hz_sq_7-28.jpg
Another set of curves at 20 Hz square and 9.8 volts
The MArinair on the left needs a little vertical adj.:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/marives_20hz_sq_9.jpg
OK, next up- phase shift. Then I have to ship this St Ives outta here.
cj
There is a big difference in the B-H curves between these two Line In transformers. I do not know how this affects the sound as I have not done a listening test on these two, but I would think you would notice something different on the bottom end.
The B-H curve is a combination between a magnetization curve and an integrated curve. They can be taken with one on the primary and one on the secondary, or both on the primary or secondary. Either way tends to give the same results, as you can saturate the core from either end of the transformer. The amount of steel dosen't change. More turns will require less current to saturate, and vice versa.
I usually use a test setup like this:
It's nice to have the scope grounds isolated by the transformer. Here is a picture of both waveforms before they are combined with the X-Y display function on the scope. The sawtooth is the integrated wave which is taken across the cap, and the other wave is the mag wave, which you might recognize from the previous post. It is taken off the resitor in series with the primary. You must adjust both amplitudes to be the same in order to get a correct BH curve. This is made easier if the scope has fine adjustments on the amplitudes:
Here is what the waves look like when they are assigned X and Y inputs on a non time domained sweep. The first rising edge is the transformer getting energized by the signal. When the core can not handle any more magnetic flux caused by an increasing voltage and thus current through the primary windings, the voltage on the other side of the transformer will no longer rise along with the primary voltage. No matter how much juice you put into the primary, the secondary voltages stays about the same because the magnetic flux in the core is at it's maximum. The rest of the energy is dissapated as a magnetic field out into free space, away from the core. The flat spot on top of the BH curve represents this maxed out voltage.
When the voltage through the primary decreases, the voltage on the secondary decreases also. But not right away. There is a slight delay, which is seen as the space between the first and second vertical lines. This is because the core is still magnetized in a North to South direction, and now you want to magnetize it from South to North. But it has a little memory which must be overcome. Reversing the polarity of the voltage does this, but with a slight delay. The energy lost in switching the magnetic flux around in an opposite direction is called core loss. This is one of the things that chews up your signal in a transformer, especially at the low end.
As the frequncy decreases, this gives the core a longer time to get magnetized. Thus, the lower the frequency, the more core loss, and therefore, the wider the area between the two vertical lines.
Here is the St Ives at 25 hz, and then 15 hz showing the increase in core loss. You have to click on these, don't want to load up the post to much:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/ives_25.jpg
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/ives_15.jpg
So it is desirable to have core material that can have it's magnetic flux switched back and forth without using up a lot of energy. You can see this loss on the following BH curve. Remember that the delay in getting the core flux reversed is what seperates the two vertical lines. So the more distance between these lines, the more core loss. In fact, you can calculate the core loss by measuring the area inside the BH curve:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/bh_demo.jpg
OK, so now that your totally confused, lets look at some Marinair vs St Ives Neve Line In transformer curves.
Here is a side by side shot of the Marinair on the left, St Ives on the right, taken with a 20 Hz sine wave at 7.24 volts on each. Don't ask me what that weird lookin curve is from on the St. Ives, as I don't know:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/marives_20hz_sine_7.jpg
Here is a side by side of a 20 Hz square wave, at 7.28 VAC rms input, which is slammin line input iron pretty hard. Not a huge difference here:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/marives_20hz_sq_7-28.jpg
Another set of curves at 20 Hz square and 9.8 volts
The MArinair on the left needs a little vertical adj.:
http://vacuumbrain.com/The_Lab/TA/St_Ives/B-H_Curves/marives_20hz_sq_9.jpg
OK, next up- phase shift. Then I have to ship this St Ives outta here.
cj
Very, very good info! Thanks!
It could look a bit like assymetric saturation (magnetized core?) combined with phase reversal.. Just a guess..
Jakob E.
It could look a bit like assymetric saturation (magnetized core?) combined with phase reversal.. Just a guess..
Jakob E.
OK, I negotiated the demise of Mr. Ives. No one gets out of here alive folks:
I re-lammed it and re did the BH curves, and have some phase shift plots also. Will pull wire off the spools tonight to confirm reverse winding or not.
cj
I re-lammed it and re did the BH curves, and have some phase shift plots also. Will pull wire off the spools tonight to confirm reverse winding or not.
cj
Ptownkid
Well-known member
So which ones were used where? Will i be happy with th st ives for som 1272's
cj wrote:
Well guess what folks, after unwinding a St Ives coil, I found that the reverse winding had indeed been skipped. Does this make it a bad transformer? No. It's just not a "Marinair" transformer.
And after re lamming the thing, the curves changed back to an original Marinair 31267.
So bottom line is - St Ives vs Green Marinair-Line In:
1) Mechanical structure exactly the same as green Marinair
2) Turns about the same
3) Reverse winding is not used
4) Lams are different
If you want to see BH curves side by side with old lams installed, let me know.
cj
Well guess what folks, after unwinding a St Ives coil, I found that the reverse winding had indeed been skipped. Does this make it a bad transformer? No. It's just not a "Marinair" transformer.
And after re lamming the thing, the curves changed back to an original Marinair 31267.
So bottom line is - St Ives vs Green Marinair-Line In:
1) Mechanical structure exactly the same as green Marinair
2) Turns about the same
3) Reverse winding is not used
4) Lams are different
If you want to see BH curves side by side with old lams installed, let me know.
cj
Flatpicker
Well-known member
[quote author="CJ"]...but the leakage inductance is different in some places makes me wonder if St. Ives for-went the flipping of the mandrels, which is a pain in the rear, but must be done if reverse winding is to be accomplished...[/quote]I may have missed this somewhere but can you (or anyone) explain more about what is accomplished by using the reverse winding? I was planning to use a St Ives as a mic input transformer as per Joe's method of reversing the primary and secondary and just using one winding of each. But of course, he used a Marinair for his test. It would be interesting to know how they would compare in this application.
Family Hoof
Well-known member
Flatpicker,
For whatever reason I found that the St. Ives half windings in reverse sounded closer to the 10468 (T1444) than a later Marinair (T1452) did, but that's just me.
For whatever reason I found that the St. Ives half windings in reverse sounded closer to the 10468 (T1444) than a later Marinair (T1452) did, but that's just me.
Flatpicker
Well-known member
Thanks, Jens - that's the sound I want. :thumb:
The lack of reverse winding seems to have brought up the leakage a bit, which would affect the high end a little. But most of the difference I saw between the Marinair and St Ives was at the bottom end due to a different lamination being used.
Similar threads
