[Roger A. Modjeski]

In recent posts the topic of who makes good output transformers has come up. I would like to hear your thoughts on this, what you may have heard in the forums and some links to transformer discussions in other forums. It appears that transformers are being assessed by the DIY community without consideration of measurements that will have significant effects on the performance of the project in which they are used. Unsupported opinion gets a bit tiring after awhile.

I have made a poll for my curiosity and will take suggestions of questions to add.

[BobRex]

A couple of questions:

1) Are you referring to manufacturers that supply transformers to the general public or those that build amplifiers around them?

2) Aren't you possibly shooting yourself in the foot by providing option 5?

[kenkirk]

Roger,
I am wondering what you think about Tamura transformers. I have always heard the vintage Sansui tube amps with Tamura output transformers were special. I also think the Hafler output transformers in the Dynaco ST 35 and SCA 35 are well matched to the EL 84's. My restored SCA 35 measured very well presenting a nice square wave at 20 hz and 20,000 hz as well. My Cat JL 1's seem to have nicely designed output transformers, but they are 55 pounds each. I wonder if that much wire effects the high frequency resolution. But the amps sound wonderful. Bass is thunderous and tight. Your thoughts?

Ken

[Clio09]

High end audio has gone off the deep end. Been that way for a while. It is one thing for an audiophile to want to think of their hobby more in terms of an art than in terms of science, after all art is subjective and relatively easier to understand than science. On the other hand it's another thing to see how many of todays designers of equipment are taking this approach as well. In my opinion many of these designers have skills that are barely above DIY level.

I have known Jack Elliano from Electra-Print for several years now. He is in the growing minority of designers who has steadfastly used measurements to support his transformer and circuit designs. I know of several occasions where people have submitted specs of output transformers they wanted him to wind along with the circuit they were going to use them in, and he refused to wind the transformers because in his opinion the specs were not correct for the circuit they would be used in.

It appears like anything else in this hobby that transformers have gone the way of capacitors, resistors, fuses, and wire. Which is to say people will use them in circuits because of their name or some myth associated with them as opposed to understanding how they will interact in the circuit.

[poseidonsvoice]

High end audio has gone off the deep end. Been that way for a while. It is one thing for an audiophile to want to think of their hobby more in terms of an art than in terms of science, after all art is subjective and relatively easier to understand than science. On the other hand it's another thing to see how many of todays designers of equipment are taking this approach as well. In my opinion many of these designers have skills that are barely above DIY level.

I have known Jack Elliano from Electra-Print for several years now. He is in the growing minority of designers who has steadfastly used measurements to support his transformer and circuit designs. I know of several occasions where people have submitted specs of output transformers they wanted him to wind along with the circuit they were going to use them in, and he refused to wind the transformers because in his opinion the specs were not correct for the circuit the they would be used in.

It appears like anything else in this hobby that transformers have gone the way of capacitors, resistors, fuses, and wire. Which is to say people will use them in circuits because of their name or some myth associated with them as opposed to understanding how they will interact in the circuit.

+100.

Best,
Anand.

[Freo-1]

I would be careful about painting all transformers from any given vendor with the same brush because one sample measured out of spec in form or another. 
 
If a high percentage of a sample block measures out of spec, then yes, there are issues.  There are other issues with some transformers, such as hum/noise emitted from the actual transformer.

[Roger A. Modjeski]

A couple of questions:

1) Are you referring to manufacturers that supply transformers to the general public or those that build amplifiers around them?

2) Aren't you possibly shooting yourself in the foot by providing option 5?

1. I am referring to all makers of transformers. I hope that companies that build amplifiers are testing what they are getting in. Saul Marantz told me that after awhile they had a rather big pile of rejected Model 8 output transformers from their regular (good) vendor. He said it was better business to not return them. They werent defective, just out of spec and not by much. He was very particuar as am I which is why we wind in house.

2. How am I shooting myself in the foot on #5

[Roger A. Modjeski]

Roger,
I am wondering what you think about Tamura transformers. I have always heard the vintage Sansui tube amps with Tamura output transformers were special. I also think the Hafler output transformers in the Dynaco ST 35 and SCA 35 are well matched to the EL 84's. My restored SCA 35 measured very well presenting a nice square wave at 20 hz and 20,000 hz as well. My Cat JL 1's seem to have nicely designed output transformers, but they are 55 pounds each. I wonder if that much wire effects the high frequency resolution. But the amps sound wonderful. Bass is thunderous and tight. Your thoughts?

Ken

I have not measured Tamura but would if someone sent me one. Sansui used Tamura because they were there in Japan. The Dynaco transformers are good, not as good as the Peerless 20-20 series which were the best I have seen. Remember that the Stereo 70 sold for $99 in a kit and the ST35 less. Back out the other parts and you will see the transformers only cost a few dollars.

I have not measured the CAT transformers but I have great respect for Ken Stevens work. I would expect them to be very good. The weight speaks to a lot of iron which is necessary for deep bass.

[Roger A. Modjeski]

High end audio has gone off the deep end. Been that way for a while. It is one thing for an audiophile to want to think of their hobby more in terms of an art than in terms of science, after all art is subjective and relatively easier to understand than science. On the other hand it's another thing to see how many of todays designers of equipment are taking this approach as well. In my opinion many of these designers have skills that are barely above DIY level.

I have known Jack Elliano from Electra-Print for several years now. He is in the growing minority of designers who has steadfastly used measurements to support his transformer and circuit designs. I know of several occasions where people have submitted specs of output transformers they wanted him to wind along with the circuit they were going to use them in, and he refused to wind the transformers because in his opinion the specs were not correct for the circuit they would be used in.

It appears like anything else in this hobby that transformers have gone the way of capacitors, resistors, fuses, and wire. Which is to say people will use them in circuits because of their name or some myth associated with them as opposed to understanding how they will interact in the circuit.

Thank you for your input. I couldn't agree more. However the transformer myth is not a wide spread as capacitor myths most likely because it is much harder to replace a output transformer for many reasons, including having to re-tune the feedback which takes equipment and a lot of knowledge. Caps don't change the feedback so their substitution is child's play. Furthermore you can see the value on the cap you are replacing and buy an identical value. Not so with an output transformer, they have too many specs to match and where are you going to get these specs anyway?

[Roger A. Modjeski]

I would be careful about painting all transformers from any given vendor with the same brush because one sample measured out of spec in form or another. 
 
If a high percentage of a sample block measures out of spec, then yes, there are issues.  There are other issues with some transformers, such as hum/noise emitted from the actual transformer.

And just how are you going to know if you get a good one or a bad one with no specs to check? Even with the specs it takes a lot of equipment and knowledge to confirm them.

[Freo-1]

And just how are you going to know if you get a good one or a bad one with no specs to check? Even with the specs it takes a lot of equipment and knowledge to confirm them.

I don't know, Roger.  How does one tell?  I can tell you a lot of DIY'ers have bought new transformers (power and output) over the years, and many of them post that their builds are successful.  My buddy who makes a lot of amps has used transformers from quite a few of the current suppliers of transformers.  He has told me that the Heyboer transformers are well made, the iron used is larger than most of the other makers for a given wattage level, and he has found them to all work well without issues. 
 
What other options are available?  I did what research I could on some of the major audio websites that support a large DIY community, and Heyboer was on of the outfits that came with high recommendations.  All I can tell you is that the results from my preamp and power amps sonically are very successful, and in fact, they sound better than a lot of commerical tube products I have owned. 
 
One of my tube dealers has a savvy tech who makes custom DIY amps, and provides technical repairs to customers in their areaa.  He told me that many of the tubes amps he has worked on from the Asian market have sub-optimum output transformers.  They measure bad from a frequency response perspective,  in that many of them start a frequency drop off above 8Kz. 
 
What about the quaity of the current production tubes?  They seem to be of marginal to poor quality.  There is no way the current 6L6/EL34's are anywhere near as good as NOS 1625/807's.  The sound quality is not that good, and longevity is very questionable.  Hence, for DIY, may as well use qulity NOS tubes, and get transformers made to support them.
 
 

[randytsuch]

I was looking for an output tranny for a tube project a little while ago, and found this post in a search
http://www.diyaudio.com/forums/tubes-valves/228625-all-dht-set-headphone-amp-51.html#post3371885

In subsequent searching, the names on the list appear to be well respected output transformer manufacturers.  Budp is Onetics.  Dave Slagle is well known for his autoformers.

From my very limited understanding core material, geometry and probably a lot more stuff factors into making a good output tranny.

Randy

[Roger A. Modjeski]

How about Lundahl?

[Freo-1]

How about Lundahl?

How about them?  Goes without saying that they have an excellent reputation.  After all, they are Thomas Meyer’s favourite source for iron.  They are also not cheap.  When I decided to make the 6AH4 preamp, the cost to get Lundahl iron was a LOT more than getting the iron made domestically.  Upon completion, the locally made iron worked very well.  It made the difference between making the product and not making the product.  The performance from the 6AH4 is remarkable, and sounds a LOT better than the commercial preamps I have owned.

I would still rather take my chances using NOS power tubes and getting iron made that supports the amp/circuit than using sub optimum current production tubes.  So far, the sonic results have been remarkable, and the power tubes are both reliable and long lasting.  I can’t say the same for the current production tubes.  The QC and reliability is just not good.  Since the tubes I have using were RF transmitter tubes, I have had to use snubber circuits to prevent oscillation.  They have worked very well, so the amps have been very stable.
 

[sfox7076]

Roger,

When did you measure a Heyboer?  It is possible that the quality has changed.

Shawn

[Freo-1]

Roger,

When did you measure a Heyboer?  It is possible that the quality has changed.

Shawn

Again, I would be careful about painting all of anything as sub optium with a sample rate of one. 

[sfox7076]

That's true.  It certainly isn't as big a sample size, but if it is also 10 years since the test, the test is even more suspect.  I suppose the test conditions and equipment are an issue as well.  The potential variables are wide.  I am exploring the underpinnings of the determination by Roger.