Heyboer and MM 18W Output Transformers

[Mike K]

I get equal turns on both legs of the primary. There is no imbalance between the reflected impedances.

Heyboer (full primary) turn ratios:

4R: 53.6:1
8R: 34:1
16R: 23:1

Merc (full primary) turn ratios:

4R: 48.4:1
8R: 34:1
16R: 23:1


I want to repeat this measurement with a better instrument. My signal generator only puts out 1.5V, so that doesn't give me much resolution on the secondary. My variac would be nice here, but dead. I'm going to see if I can find a 15-20V transformer that I can use.


If the winding ratios are the same, which no doubt they likely are. We'd have to assume from the DCR that these are both wound very differently on both the primary and secondary.

[Helmholtz]

If your generator has low enough output impedance you can feed signal to the secondaries.
Otherwise use heater voltage to secondaries. Expect up to 300V at primary.

[Mike K]

If your generator has low enough output impedance you can feed signal to the secondaries.

No - doesn't work, I tried.


I calculated the efficiencies of each for middle frequencies at 8 ohm. The Heyboer is more efficient, mainly due to the low secondary R.

Heyboer = 88.5% (Rp = 618R, Rs = 0.5R, n = 34)
Merc = 84.5% (Rp = 533R, Rs = 1.0R, n = 34)


If this is a volume contest, I think we'll know who wins provided there isn't something hiding in the core.

[g1]

I could buy an inexpensive one but they all seem to have limitations. Some only measure at 1kHz. From what I see, a decent handheld is around $300. I've never had the need for one, so I'm not sure it's a great investment for me.

DE-5000 is a highly regarded LCR for the cost, you can pick them up for around $150.

[Mike K]

DE-5000 is a highly regarded LCR for the cost, you can pick them up for around $150.

Thanks for the tip. I was going to perhaps ask in another thread. Might be a better tool to have than a variac.

[Mike K]

I found a little more information on the Heyboer. Apparently it is M27 steel in the core. That can be purchased in sheets of 0.014", 0.0185", etc... I count 68 laminations at a distance of 0.97", which gives a thickness of 0.014" per.

Specifications for that material can be found here:

https://www.brown.edu/Departments/En...ery%20good.pdf

[Mike K]

I found some interesting comments from Dan Boul, who spent a lot of time and money developing his own transformer design with Merc for his company, 65 amps. Apparently they worked some with the RS 18w but I think ultimately decided to go with something that is more built around the RS De Luxe, I believe, but modified to their taste.

Here's what he has to say and outlines some key differences between what Heyboer interpreted and what they did. They chose to stick with a very vintage and expensive steel spec that more closely replicates the original, specific interleaving and winding patterns that they found from their favorite samples.

Regarding Radiospares, that's another whole can of worms. We found several original Radiospares Deluxe EL84 OT's (which is the original Marshall one) and Mercury had a couple they had rewound and we found out that all of them are different. We actually found an NOS OT that was the same design as one our our used ones and found out on Mecury's computers that they had aged extremely differently as you could imagine. They did NOT sound the same at all. This illustrated just how much of a wild guess it can be to recreate a 40 year old sound. I'm sure if we found other ones that had aged differently, we would have seen different results. So how do you know if you're getting a correct reproduction? You're not. From anyone. Period.


As I mentioned, there is no single "Radiospares" design. Some were very simple with machine wound patterns inside and others were very complex such that they must be wound by hand. Some had heavily laminated stacks, others did not. It seems that RS changed that OT almost every year.

With that being said, the 18watt tranny that Mercury offers is correct in a couple of ways that I think are unique to Mercury. Most importantly, the steel is time-correct for the 60's. They use hand-annealed steel that is cooked by hand in the 60's and so does Mercury (this is the main reason they cost so much), not modern silicon-steel which definitely affects tone and reaction. For the $20k hi-fi amps they supply the trannies for, there is no substitute. Second, 18watt RS OT is based on one of the complex RS trannies that they felt sounded the best. It had 9 interleaves which definitely increases tonal range and reduces the "buzz" factor. 9 interleaves dictates that is must be hand wound, another reason they cost so much.

https://www.18watt.com/viewtopic.php?t=7201

[loudthud]

Can "They" artificially age or burn-in new transformers to duplicate the sound of the old used ones ?

[Mike K]

Can "They" artificially age or burn-in new transformers to duplicate the sound of the old used ones ?

I don't know? Ask "them".


With regard to these, I'm pretty sure both Heyboer and Merc tried to make them sound like they would when they were new. The information I found is that Paul Ruby helped develop the core specification for the Heyboer. The rest of it is based on some version(s) of vintage designs - perhaps machine wound versions, because I believe they only machine wind.

The current Merc is based on that 9 interleave and vintage iron spec as quoted above. That's why (even with the best pricing) the Merc costs $170 and the Heyboer $70. If you were trying to make money building these things, there's a clear choice where you might go. Heyboer can produce quantity, they are less than half the cost and they reportedly sound very good, perhaps better than the very expensive, complex design to some people. If you're building one or two amps, and you want to make it the best for you or your customer's purposes. It's probably best to try both. $100 isn't all that much in terms of tube amps and it's easy to resell (or return - you can, and I have, returned Mercs with uncut leads for full refund).


As far as the "aged" components debate. I had a guy contact me about an amp and give me the run around about that. He asked me if I thought there was a difference in aged components. Yeah, I said, probably. But all those recordings you love and base your opinion on weren't made with aged components. All that stuff was new or typically a decade old, at most, when that was done. Other than that, they are just a liability. Do what you like, but my philosophy is to try to use what was learned from 70 years of tube amps, not try to replicate what 70 years has done to them.

There are a number of pros and aficionados using vintage amps these days, but when I see the guts, not much remains of the original. Typically an original, but reconed speaker. Maybe the original transformers, but a lot of times the OT has been replaced. Caps are usually not original, and if they are, probably should be replaced. Maybe a few original carbon comp resistors that haven't turned to snakes...

[Chuck H]

... my philosophy is to try to use what was learned from 70 years of tube amps, not try to replicate what 70 years has done to them.

This^^^^^^^^^^^^^^^^^^^

I have had run ins with with people over this. Mostly players, but there was a "tech" in my area for a time with some weird philosophy on the matter too.

[Chuck H]

Oh... Regarding the transformer matter (just to get closer to the actual topic )

If the goal is to get as close as possible to the original 18W Marshall, well, fine. If the goal were to just find the best SOUNDING transformer for an 18 watt type build I think the Hammond 1608 should be included for evaluation. Though obviously the footprint doesn't match. Still...

shea, a member here we haven't heard from in twelve years put me onto this "off the shelf" Hammond model when he made a splash in an amp shootout with his 18W builds using this iron. My own personal amp uses one. I've built the same design with two other brands, though Mercury was not one of them Heyboer was and Classictone the other. Hands down "I" like the 1608. By a lot. I quoted "I" because tone is subjective so this can only be an opinion. Also, my amp isn't an "18W" properly. But it is a 2xel84 designed to be cranked for distortion. shea did show that in an amp shootout this OT can also sway a panel, so.?.

The Heyboer was good. The Classictone did nothing for me and I thought it sounded anemic with limited range at both ends of the audio.

This, FWIW considering I haven't offered anything regarding a purpose made "18W" OT.

[Mike K]

Oh... Regarding the transformer matter (just to get closer to the actual topic )

If the goal is to get as close as possible to the original 18W Marshall, well, fine. If the goal were to just find the best SOUNDING transformer for an 18 watt type build I think the Hammond 1608 should be included for evaluation. Though obviously the footprint doesn't match. Still...

shea, a member here we haven't heard from in twelve years put me onto this "off the shelf" Hammond model when he made a splash in an amp shootout with his 18W builds using this iron. My own personal amp uses one. I've built the same design with two other brands, though Mercury was not one of them Heyboer was and Classictone the other. Hands down "I" like the 1608. By a lot. I quoted "I" because tone is subjective so this can only be an opinion. Also, my amp isn't an "18W" properly. But it is a 2xel84 designed to be cranked for distortion. shea did show that in an amp shootout this OT can also sway a panel, so.?.

The Heyboer was good. The Classictone did nothing for me and I thought it sounded anemic with limited range at both ends of the audio.

This, FWIW considering I haven't offered anything regarding a purpose made "18W" OT.

No - that's not the goal. I'm not sure I know what an original 18W Marshall sounds like. I just plan on using (half) that circuit because it's easy to build to test. As far as what I'll actually build as some final product with either (or both) of these OTs isn't certain. If it turns out I really like how the 18W circuit turns out, I may stick with that and add something on the other side of the PI like TMB channel or EF86 channel and build it into a real chassis (my proto chassis is like swiss cheese).

I was actually looking at the specs of the Hammond last night and had the same idea. Actually I was looking at the 1750PA which is the 18W OT. But I also came across the 1609 and that certainly would be a contender. If I was doing a true "shootout" I'd go for it. But as of now I think I'm just going to compare these two. I have a lot of parts now as it is so I think I'm going to hold off on spending anymore, particularly with the prices the way they are, and work with what I have. I'm sure I'll have at least one of these OTs on an amp in the next year or so, so perhaps when I have some funny money to buy more parts, I'll do a comp with the Hammond.

Right now my goal is just to get myself motivated to work on this and perhaps learn something, if not then entertain someone and show there are real measurable differences between these things and real audible differences in the end. I have no intention of proving one is better than another. That's like trying to prove caviar is better than chicken eggs. Both eggs, one is more expensive and both taste quite different, but obviously some people are going to stick with one or the other. Personally, in this instance, I'll stick with chicken eggs

[Chuck H]

All good. Just like you're a proponent for the MM iron (and for good reasons) I happen to be a proponent for the Hammond 1608 so I mention it whenever the topic even vaguely allows.

[Helmholtz]

If this is a volume contest, I think we'll know who wins....

A difference of 0.17dB.
You must be joking.

Interestingly, assuming the turns ratio of 34.1 into 8R is correct, both OTs would have a theoretical Raa of 9.3k
Adding effective DCR gives an effective Raa of 9.9k for the Heyboer and 11k for the Mercury.

[Mike K]

All good. Just like you're a proponent for the MM iron (and for good reasons) I happen to be a proponent for the Hammond 1608 so I mention it whenever the topic even vaguely allows.

It's not out of line. I appreciate the insight and I'll give it a try in the future. Right now I'm trying to decide if I really "NEED" that $150 LCR meter.