Peerless S-217-D DIY

[MatthisD]

somehow i had the idea that this was a nickel core, who started that rumor? 

if you squint real hard you can see the grain oriented  sideways to the lam,

we went with a lace 1 X 1 core for max henries, orig is probably like the 16402,

which uses like a 5 x 5 for a little air gap action,

Peerless catalog says that this thing can be used for SE of P-P,

i am guessing the 5 x 5 lace up is for SE operation,

the Pultedc is Push Pull so we can skip the 5  x 5,

You're joking! it seemed to make sense that it was high Nickel, 436c limiter has no output attenuator which I thought might have been to keep the level on the primary within the rated max of the 16402.

Isn't the '0' in the catalog the rated unbalanced DC in mA...wouldn't the 5x5 stacked lams reduce low frequency distortion but roughly halve inductance.
Is one of the secondaries the only reverse winding in this case?

Delighted to see it finished, let us know how it goes.

 

[CJ]

i was wondering about the DC also,

they must mean  SE para-feed circuit, where the current runs thru the choke,

and a cap feeds the primary which blocks DC,

Jackies, i have the HS 52 info, how many do you need?

 

[CJ]

rumor has it that Peerless did some "borrowing"  from Western Electric in order to produce wonderful transformers,

such things as coil geometry and lams,

well, every engineer worked for WeCo at some point in their life,

here is a pic of a 12 L lam next to an old WE xfmr,

notice the ceramic type potting compound,

only stuff that won't melt with a Wagner heat gun,

had to use a hammer,

 

[jackies]

Jackies, i have the HS 52 info, how many do you need?

Well, I was hoping I could get my hands on a pair - in case I ever gather myself to begin the 670 build.
I was trying to bid on them on ebay last week - needless to say I was pathetically outbid, they went for $1200 something. Other options would include purchasing Sowters or winding something, whereas I'm, of course, concerned about primary inductance - have you measured what original HS-52's primary inductance is like?
Sorry for the offtop.

 

[CJ]

we can do the Peerless and the Triad on a custom  C core and get better results,

here  is a graph of the Peerless S-217-D Inductance vs Frequency

calculated from input current with no load on the secondary,

perm is proportional to inductance, so you could call it Perm vs Freq if you like,
we maxed out at 220 Henries at 30 Hertz, which equates to a perm of 10,000

 

[CJ]

Type 12 L Lamination Dimensions

had some friends in South America requesting this info

so that they can get the mojo workin down there,

Cross Section Area should be exact,

Magnetic Path Length-L was done with a piece of solder bent around the diagram,

do this about 2/3 of the way out, not in the center,

stack used in the S-217-D is 7/8th inch,

about 0.52 lbs worth,

 

[lassoharp]

whereas I'm, of course, concerned about primary inductance - have you measured what original HS-52's primary inductance is like?

I have measured the pri L on my HSM-82 - 18.38H  @ 120Hz.  Triad claims +/1 db from as wide as 7Hz-50K when used at 1/4 max power level (rated at 15W).  20Hz-20K if slamming it full.

I may be jumping to conclusions but I'm aware of the recent thread comparing the L of Sowters recreation of UTC A-26 and the measured L of the original and the LS-51 - those crazy high L numbers.  Something peculiar to UTCs design that gives the high numbers?    Triad HS/M -82 is 8K:500 vs 30Kish:500 of the UTC PP. 

 

[jackies]

Nothing peculiar, high nickel permalloy and lots of turns gives high inductance, and regular steel with fewer turns gives lower numbers.

 

[CJ]

i am finding out that the standard lam that UTC uses is not 80% Ni,

more like a 50/50, perm comes out to 20,000 when i do the math for different models,

pri inductance depends a lot on the turns ratio,

the S-217-D is 4.5:1

UTC A-26 is 7.5:1

if both terminate to 600 ohms, the pri Z will be

12K and 30K

so you need more Henries for the 30K to get the same bass roll-off point,

but inductance is proportional to turns squared, so it goes up quick with added turns,

UTC has tons of turns, but the tradeoff is DCR,

here is the math on the A-26> new format, save worksheet to txt, 

UTC A-26 Magnetics
Max dbm: plus 15
Power Level: 30 mw Bass Frequency: 20
Unbal. DC: 0 Operating Level- Flux:5270
Frequency Response: 2db:  20 - 40K
VAR: 0.030
Lamination: 31 UI
Area-cm^2: 6.3000000E-01
Area-in.^2: 0.097650
Former: paper

Multiplier: 1.6 Stacking Methog: 1 x 1 B= 1240079 *E/N
Total Area-cm^2 1.01 Total Area-in^2 0.1562 Lam Count: 37
x 0.014= 0.518 in.
MPL-cm. 9.53
MagMet Ind Coef: 8.322E-02 Henries for -3db rolloff: 277.3 Hertz: 20
Pri Turns 8000 Pri. Ind: 1533.911 R-pri 2660
Pri Inductance with Gap: 1531.193 Turns/Volt: 235.2941
Turns Ratio: 7.62
X-L pri 192416
Sec Turns 1050
Sec Ind: 26.424 R-sec 89 Sec Inductance with Gap: 26.377 X-L sec 3314.66
Wire Size: 0.000477
Pri Z-Ohms 34830 Z Ratio 58.0 XL-Pri: 192757 Volt-Turns 0.004250
Q-Pri: 72.5 Z root: 7.6
Sec Z-Ohms 600 XL-Sec: 3321
Turns 8000
Q-Sec 37.3
Alloy Perm: 20,000 Effective Perm: 20000
L-gap in: 0 L-gap cm: 0
Voltage Loss: 1.1973 Optimun gap: 0.0012 mils
Insertion Loss-db: 0.7821

opt. gap: (.4piN^2Ac^-8/L)
Volts Pri-RMS: 34
Volts  Sec RMS: 4.463 15.490 VA: 0.0000
DC-balanced: 0.009 Amps
VAR: 0.03
DC-unbalanced: 0
H-constant: 0.13200 Oersteds: 1.05600
Flux-AC-pri: 5270
Flux-DC-pri: 0

I-ac= 0.0002 Amps-rms @ Lowest f
Idc + Iac 0.0002 Amps

 

[CJ]

almost ready to fire this thing up, 


check the funky foil coupling caps from 1887,

check the jb weld on the wima caps,

primitive sheet metal clamp soldered to chassis, aviator shears, $17.99

the green resistors can be found on ebay,

i hope i connect everything correctly so as not to blow up the xfmr, 

 

[CJ]

here is the leading info on the peerless s 217 d,

this is how the flipped coil was verified,

blu and blu wht will be the same place on each coil,

so if one is longer than the other,

then the coil is flipped.

these are not the leads that you wire to,

they are internal leads that go to the weird header with the 4 terminals in the middle,instead of the normal pinout

that is why you see the 1,1  2,2  etc.

 

[buildafriend]

I would think anything hockey puck like would just make a Canadian happy.

Thanks for sharing, I can't wait to view this on a computer instead if my tiny phone. Did you use your DIY coil winder for this? I love your turns counter on that thing.

 

[CJ]

yeah, diy winder, #29 and #36 pretty easy,

 

[CJ]

finally wired up the output xfmr,  sounds great!

had some problems with the line amp oscillating, but after the feedback windings were swapped, most of it went away, 

just occasional motor-boating at certain settings, so i shortened up some grid leads,

changed the bypass cps to .33 instead of .68 to swamp out the low end,

still had a slight problem, so i added the stock zobel, 62k/.003  to the 600 ohm output winding and that fixed everything.

i am wondering if there were stability problems with the line amp/xfmr combo,

there is a 20 pf across the plates of the 12ax7a on the original circuit,

this is a fix for an unstable circuit, or it might just mean wiring capacitance,

i also noticed that when the output was run balanced, (ground lifted) that the zobel was no longer needed.

i am also running the orig zobel on the hs-29, 63k/270pf, and using the 620 ohm resistor across the triad hs-56 sec.

you want that resistor in there so the line amp does not distort, the level goes up quite a bit without it, it seems to make no difference on the high end,

tons of bottom end, very happy with the sound, no hum, no tube noise,

i took out the bandwidth control as it seems like a "bells and whistles" thing,

it only really works at certain settings, and i have enuff controls to hassle with if you include the mic pre, compressor, etc.

i did not need the zobel with the Sowter output to fix any problems,
so Brian might have done a better job at controlling phase shift on the tertiary winding,
he uses a EI core instead of a dual coil L lam setup, flux linkage with the EI can sometimes be better than the UI or L dual coil setup, opposite of what you might think because the dual coil wraps more of the core, leakage causes phase shift which causes feedback,

a 0.003 cap is a pretty big value to strap on the output xfmr, this might be the fix for the phase shift.

this is a unique line amp when you look at it, cathode current for V1 goes thru the output xfmr,

there are 4700 ohm  self bias cathode resistors on the 12AU7, which elevates the cathode to 40 volts, which cuts off the tube,
but they elevate the grids to about 30 volts with a voltage divider off the pwr supply, to get the operating point back up, i have about -8 on the 12AU7, slightly lower due to a dropped B= voltage, (240 VDC)
this is probably done to make the amp less noisy,
or it could be to limit current thru the transformer pri, a 12AU7 can pump out about 20 ma, so 40 ma might be too much for the pri wire, too much heat, pri dcr is about 395 ohms, times 40 ma = 15 watts!
with the 4700, we get about 8.8 ma times 2 = 17 ma times 395= 6.75 watts,

ok, bolt the lid on and rack this so we can finish the 670 and fender princeton and v72  pwr trans revision,

i recommend mounting the NOS Triad and Peerless stuff flat, rather than using the original hardware as these are wax filled units, and sometimes, when they solder the header plate on, the wax can get hot and boil up into the seam, so you have a contaminated weld,

you can see this over at gearslutz, people asking how to fix their 2000 dollar output transformer that is hanging by the wires which may pull out,
so yeah, a real bummer to finally score a s 217 d and have the leads ripped out,  :-\

if you own an original EQP-1a, then you may want to reinforce the transformer mounts with some type of strap over the top,

so you can fix tape hiss, jack up the treble 8 different ways,and best of all,
boost and cut the bass at the same time for that pultec mojo,
we have a Telefunken 12AX7a and a GE ECC/12AU7a for the tubes, offboard pwr supply,

now for vocals, nuthin beats the MEQ-5.
i would like to use this line amp with a meq 5.

this chassis is what i would call a "white trash" chassis, kind of rough, done in a hurry,
it is a rebuild of an earlier chassis, we try to be neater now days as there seems to be some real sharp projects coming down all the time,

 

[radiance]

That's friggin awesome CJ! Love it....

 

[opacheco]

finally wired up the output xfmr,  sounds great!

had some problems with the line amp oscillating, but after the feedback windings were swapped, most of it went away, 

just occasional motor-boating at certain settings, so i shortened up some grid leads,

changed the bypass cps to .33 instead of .68 to swamp out the low end,

still had a slight problem, so i added the stock zobel, 62k/.003  to the 600 ohm output winding and that fixed everything.

i am wondering if there were stability problems with the line amp/xfmr combo,

there is a 20 pf across the plates of the 12ax7a on the original circuit,

this is a fix for an unstable circuit, or it might just mean wiring capacitance,

i also noticed that when the output was run balanced, (ground lifted) that the zobel was no longer needed.

i am also running the orig zobel on the hs-29, 63k/270pf, and using the 620 ohm resistor across the triad hs-56 sec.

you want that resistor in there so the line amp does not distort, the level goes up quite a bit without it, it seems to make no difference on the high end,

tons of bottom end, very happy with the sound, no hum, no tube noise,

i took out the bandwidth control as it seems like a "bells and whistles" thing,

it only really works at certain settings, and i have enuff controls to hassle with if you include the mic pre, compressor, etc.

i did not need the zobel with the Sowter output to fix any problems,
so Brian might have done a better job at controlling phase shift on the tertiary winding,
he uses a EI core instead of a dual coil L lam setup, flux linkage with the EI can sometimes be better than the UI or L dual coil setup, opposite of what you might think because the dual coil wraps more of the core, leakage causes phase shift which causes feedback,

a 0.003 cap is a pretty big value to strap on the output xfmr, this might be the fix for the phase shift.

this is a unique line amp when you look at it, cathode current for V1 goes thru the output xfmr,

there are 4700 ohm  self bias cathode resistors on the 12AU7, which elevates the cathode to 40 volts, which cuts off the tube,
but they elevate the grids to about 30 volts with a voltage divider off the pwr supply, to get the operating point back up, i have about -8 on the 12AU7, slightly lower due to a dropped B= voltage, (240 VDC)
this is probably done to make the amp less noisy,
or it could be to limit current thru the transformer pri, a 12AU7 can pump out about 20 ma, so 40 ma might be too much for the pri wire, too much heat, pri dcr is about 395 ohms, times 40 ma = 15 watts!
with the 4700, we get about 8.8 ma times 2 = 17 ma times 395= 6.75 watts,

ok, bolt the lid on and rack this so we can finish the 670 and fender princeton and v72  pwr trans revision,

i recommend mounting the NOS Triad and Peerless stuff flat, rather than using the original hardware as these are wax filled units, and sometimes, when they solder the header plate on, the wax can get hot and boil up into the seam, so you have a contaminated weld,

you can see this over at gearslutz, people asking how to fix their 2000 dollar output transformer that is hanging by the wires which may pull out,
so yeah, a real bummer to finally score a s 217 d and have the leads ripped out,  :-\

if you own an original EQP-1a, then you may want to reinforce the transformer mounts with some type of strap over the top,

so you can fix tape hiss, jack up the treble 8 different ways,and best of all,
boost and cut the bass at the same time for that pultec mojo,
we have a Telefunken 12AX7a and a GE ECC/12AU7a for the tubes, offboard pwr supply,

now for vocals, nuthin beats the MEQ-5.
i would like to use this line amp with a meq 5.

this chassis is what i would call a "white trash" chassis, kind of rough, done in a hurry,
it is a rebuild of an earlier chassis, we try to be neater now days as there seems to be some real sharp projects coming down all the time,

Dear CJ,

Do you have an of those fantastic transformers drawing for this S-217-D like your bisected specimens technical drawing?
I am thinking to rewinding one bad unit that I have.

Thanks
Opacheco.

 

[CJ]

never had a real one to steal data from, just speculation by way of certain documents and a phone call from Winston O Boogie.

you could help put the final touches on this puzzle if you are careful,

would love to see some pics, rumor has it that one of the coils is upside down,  or flipped around , don't know which,

 

[opacheco]

never had a real one to steal data from, just speculation by way of certain documents and a phone call from Winston O Boogie.

you could help put the final touches on this puzzle if you are careful,

would love to see some pics, rumor has it that one of the coils is upside down,  or flipped around , don't know which,

I am thinking in repair it by myself and I would like to share my findings here but I would like your comments if this bisection procedure have any special procedure in your experience??...

Thanks
Opacheco

 

[CJ]

you are going to need a hot air gun or toaster oven to heat the wax up but first you have to get the lid off,

look for some small crimp welds between the lid and the can, you can stick a very small screw driver or knife blade in between the can and lid to break the weld,

after that heat it up a bit and try to pry the lid off,

we would be glad to do this for you, free of charge if you want to mail it up here,

after you get the transformer out, take a pic and write down where all the wires go.

you might just have a broken wire from the coil to the lid, if so, no further dis-assembly will be required.

just solder the wire.

 

[opacheco]

you are going to need a hot air gun or toaster oven to heat the wax up but first you have to get the lid off,

look for some small crimp welds between the lid and the can, you can stick a very small screw driver or knife blade in between the can and lid to break the weld,

after that heat it up a bit and try to pry the lid off,

we would be glad to do this for you, free of charge if you want to mail it up here,

after you get the transformer out, take a pic and write down where all the wires go.

you might just have a broken wire from the coil to the lid, if so, no further dis-assembly will be required.

just solder the wire.

CJ,

Thanks a lot for share your instructions and I so sorry but I live in Honduras and shipment charges to USA is very high cost for a transformer like this...anyway I would like to do that for myself, I have a little modest experience in windings.

Thanks a lot again!
Opacheco.