Restoration & upgrade of 1960s Markel Heetaire space heater

[iroc86]

Here's an unusual project for you all. A while back I picked up a few vintage Markel "Heetaire" space heaters. The design dates back to at least the 1960s. This product is actually still sold under the TPI brand, but it's now made overseas and (apparently) the quality has suffered. The original Markel heater has a reputation for being well-made, reliable, and basically indestructible.

This thread will document the restoration of this classic heater. While the underlying electrical design is quite simple, there are a number of safety and performance improvements I'm planning to make--this was, after all, designed over fifty years ago and we do things a little differently these days! This project probably won't approach the level of sophistocation that I often see on the EEVblog forums, but I hope it's interesting to at least a few of you.

The case is stamped metal and has a large, convex grate on the front. This was before the days of "tip-over" safety switches, so the grate is designed to keep the heater from completely melting the floor if it falls over. The three-point rubber mounts are a nice attention to detail and ensure the unit sits stable on the floor. I believe the "867" stamping on the bottom is a date code (8th week 1967). I have another model stamped with "1972."





Inside, things are a bit scary, though not unusual for appliances of this vintage. Point-to-point wiring and no earth ground. The plug is also unpolarized, which creates an interesting situation for safety--depending on the orientation of the plug, the thermostat could act as a "hot side" switch. There's a 50/50 chance that the heating elements will be live when the unit is cycled off via thermostat. Considering the relatively small clearances between the element coils and the case/reflector panels, this is quite dangerous if a short ever develops. Fortunately, the main on/off switch actually switches both input AC lines, so there's less worry if the heater is plugged in and turned off.

This particular unit also suffered from a non-functioning fan motor. The fan is attached to the spindle by friction fit with a rubber grommet. Over the last fifty years, the rubber hardened and caused the fan to slide down the shaft, allowing the fan blade to come into contact with the coil winding. The stalled situation didn't hurt the shaded pole motor, but I will need to fix the fan.

More to come!

[SilverSolder]

It looks in really good condition already, nice find!

[james_s]

I have a heater almost identical to that except mine is bluish, found it in my partner's grandfather's house after he passed away. The fan was stuck when I got it but otherwise it's fine, I left it completely original, I don't leave space heaters plugged in when nobody is around anyway so I'm not worried about the safety aspect.

[iroc86]

Thanks, SilverSolder and james_s! Markel seems to have made quite a few of these over the years; the earliest ones I've seen are indeed blue and have big toggle switches for the power and fan.

The first step was to take apart the switches and check for any contact damage. These looked fairly good, but I'll burnish the contacts for good measure. No real surprise due to the age and the load that is being switched (1650 W total output @ 120 VAC).





The heating elements are made of nichrome wire wrapped around a threaded ceramic rod. Markel calls this "Neo-Glo" and was actually issued a patent for the technology in the 1930s. The older (presumably domestically-manufactured) models promised a lifetime guarantee on the heating element.





Disassembled and ready for cleaning and painting!

[james_s]

This is the one I have, it's branded differently but looks identical. I don't know how old it is but I assumed 1960s, I have not yet done any cosmetic restoration.

[iroc86]

That's pretty cool, james_s. Your unit is probably a little older than the one I have. Maybe there's a date code sticker on the bottom?

I'm finishing up the cosmetic stuff... new paint, hardware, and grille. I'm fairly certain that the fibrous white insulation panel in the first photo below is made of asbestos.

I also decided to remake the front decal. The original silk screening was starting to lift off, and the poor condition wouldn't really match the rest of the refurbished unit. I patterned the artwork off the "late-model" version of this heater, but made a few visual tweaks. My wife also insisted that I retain the vintage Markel logo instead of the modernized version . I did the artwork in Adobe Illustrator and had it printed by a sticker manufacturer.

The next step will be electrical upgrades and rewiring. I'm also planning to include a small PCB, so we should be seeing some electronics stuff soon!

[SilverSolder]

Very cool!   

What company did you use for manufacturing the sticker?

[iroc86]

Very cool!   

What company did you use for manufacturing the sticker?

Thanks! I went with Sticker Mule. One of my work colleagues uses them for making promo stickers, and it seemed like an interesting idea for custom panel labeling and whatnot. I used to have access to an industrial laser engraver at my former employer, so I have to improvise now, haha. I'm quite satisfied with Sticker Mule, especially with the cost--about $1 each for a sample run of 15 stickers.

[SilverSolder]

Very cool!   

What company did you use for manufacturing the sticker?

Thanks! I went with Sticker Mule. One of my work colleagues uses them for making promo stickers, and it seemed like an interesting idea for custom panel labeling and whatnot. I used to have access to an industrial laser engraver at my former employer, so I have to improvise now, haha. I'm quite satisfied with Sticker Mule, especially with the cost--about $1 each for a sample run of 15 stickers.

Ah, that's an interesting approach.  Do you cut the internal holes yourself, or is that something you can ask the Mule to do for you?

[iroc86]

Ah, that's an interesting approach.  Do you cut the internal holes yourself, or is that something you can ask the Mule to do for you?

I used to have a punch and die set for just that sort of thing, but I'll probably just trim the holes with a hobby knife. The screws and knobs should cover up any rough edges on the cuts.

I believe Sticker Mule does perform internal cuts, but they charge extra and I think you need to do a whole production run. I tried to get them to cut the holes for me on this label, but they'd only do the outside profile--although that can be any shape, as it's CNC knife-cut.

[TimFox]

Very cool!   

What company did you use for manufacturing the sticker?

Thanks! I went with Sticker Mule. One of my work colleagues uses them for making promo stickers, and it seemed like an interesting idea for custom panel labeling and whatnot. I used to have access to an industrial laser engraver at my former employer, so I have to improvise now, haha. I'm quite satisfied with Sticker Mule, especially with the cost--about $1 each for a sample run of 15 stickers.

I went to the website, but couldn’t find anything for less than 50 units.  How did you get an order for 15?

[SilverSolder]

Very cool!   

What company did you use for manufacturing the sticker?

Thanks! I went with Sticker Mule. One of my work colleagues uses them for making promo stickers, and it seemed like an interesting idea for custom panel labeling and whatnot. I used to have access to an industrial laser engraver at my former employer, so I have to improvise now, haha. I'm quite satisfied with Sticker Mule, especially with the cost--about $1 each for a sample run of 15 stickers.

I went to the website, but couldn’t find anything for less than 50 units.  How did you get an order for 15?

Click on the "Get Samples" button on their main page.  (Yes, that had me confused for a while too!)

[SilverSolder]

Ah, that's an interesting approach.  Do you cut the internal holes yourself, or is that something you can ask the Mule to do for you?

I used to have a punch and die set for just that sort of thing, but I'll probably just trim the holes with a hobby knife. The screws and knobs should cover up any rough edges on the cuts.

I believe Sticker Mule does perform internal cuts, but they charge extra and I think you need to do a whole production run. I tried to get them to cut the holes for me on this label, but they'd only do the outside profile--although that can be any shape, as it's CNC knife-cut.

In your picture, you can see where the holes are meant to be -  but they aren't actually cuts, I guess, but thin white lines?

[iroc86]

Click on the "Get Samples" button on their main page.  (Yes, that had me confused for a while too!)

And me as well! I also couldn't figure out how to get samples the first time around. Figures it's the big orange button on the front page.

In your picture, you can see where the holes are meant to be -  but they aren't actually cuts, I guess, but thin white lines?

Yeah, those are just guidelines for the hole cutouts--knobs, screws, and a neon indicator light.

[iroc86]

When I first assessed the condition of the heater, I knew I didn't want to reuse the original fan motor. It's over fifty years old and has seen better days. The bearings are noisy and the coil wiring looks a little scary by today's standards. It may very well last another fifty years, but it didn't make sense to upgrade everything else and keep this old motor.





Fortunately, the design and mounting configuration of these little consumer appliance shaded pole motors hasn't changed in half of a century. It turns out that a bathroom exhaust fan motor is basically identical, save for one key issue: it operates at a much faster speed, about 3000 rpm versus 1500 rpm compared to the original motor. The increased speed is way too fast for a space heater--not only will it sound like an airplane engine, but it'll cool off the heating element instead of convecting the heat into the room. And this is where this project expands in scope!



I didn't have much luck finding a lower-speed motor in the same frame style, so I opted to use the inexpensive bathroom fan motor with a custom triac circuit. Most of my electronics tinkering has been with DC stuff, so this seemed like a good excuse to learn more about AC circuits (at line voltage, no less!) . This isn't the first time I've worked with mains, but never in the development of my own stuff.

I put together a basic triac circuit in LTspice based on the common wall dimmer switch. The load is resistive, which is far different than the inductive load of the fan, but it was close enough to start simulating. I also built up the circuit on some perfboard to compare theoretical and actual behavior.





The results were pretty promising. My simulation was nearly spot-on with the real circuit, demonstrating the usual chopped AC waveform.



Of course, in practice, this circuit needs some additional features for an inductive load. The main issue is dV/dt when the triac commutates. Using a motor as the load, the voltage and current are out of phase, so the voltage across the device will be very high when the current is zero. If dV/dt is too high, the triac may self-trigger.

Again, this isn't really an issue for the resistive load in the test circuit above, but you can see the behavior below. The dV/dt here is about 500 V/µs, which exceeds the 200-400 V/µs rating of the BTA08 triac I'm using.

More to come soon... an upgraded test circuit and some snubbing.

[SilverSolder]

I would have probably just used the old motor (maybe new bearings if noisy), it seems in pretty good condition.  You'd be surprised how long that old stuff can last...  it will likely outlast any electronics you put together, LOL! 

For safety, maybe find a way to ground the case, and use a 3 prong plug, and a GFCI relay somewhere?

[james_s]

I see absolutely nothing wrong with the old motor, why on earth would you replace it? If the bearings are noisy just oil them, it'll probably run for another 50 years. The coil doesn't look scary at all, it looks like it's in very good condition.

It's almost certain to be much higher quality than any of the Chinese crap on the shelves today. That's not an upgrade, it's a downgrade, with an added hack to make it work.

[iroc86]

...Chinese crap on the shelves today. That's not an upgrade, it's a downgrade, with an added hack to make it work.

Ouch!

Earlier up the page I mentioned that I have several of these heaters to refurbish, so swapping the motor is a universal solution I can apply across all of the units. The perfboard test circuit may appear as a hack, but it's not representative of the final design.

I used to live in China. I saw all sorts of hokey products, especially in their domestic market. I understand where you're coming from by suggesting the reuse of the potentially "higher quality" original parts, but in the case of the fan motor, it's not something I'm interested in doing. Besides, this is just a fun side project; if my design choices allow me to explore some new electronics concepts, then all the better... right?

For safety, maybe find a way to ground the case, and use a 3 prong plug, and a GFCI relay somewhere?

Yup, good call--that's in the plans. Case grounding, an inline circuit breaker, and wiring up the second heating element to shut off if the motor fails (since the fan serves to cool the second 750 W element). Might also add some filter caps to cut down on noise from the triac.

[Prehistoricman]

The increased speed is way too fast for a space heater--not only will it sound like an airplane engine, but it'll cool off the heating element instead of convecting the heat into the room.

What do you mean by cooling the element off? That's how heat is transferred...

[iroc86]

What do you mean by cooling the element off? That's how heat is transferred...

Sure, I can clarify what I meant. It's a trade-off between temperature and flow. The higher fan speed would blow heat farther into the room, but the outlet temperature will be lower due to the higher air velocity across the element. The fan speed in the Markel unit is just fast enough to help direct heat from the rear element out the front grille, not to really aid in room circulation.

Also, the second (rear) element is arranged back inside the case, rather low. It doesn't have a full concave reflector like the front element. The rear element only operates with the fan, because it'd otherwise just radiate heat on to the backside of the front element. If the fan ever failed (as it did in this particular unit; see post #1), the unit will get extremely hot. So, I'm including a safety feature to turn off the rear element if the fan somehow shorts out or stops working.

[SilverSolder]

Changing the motor is an age thing, I think...   once you have enough projects under your belt, you learn to "leave well enough alone" because the time saved means you can do even more crazy projects!   

[iroc86]

Changing the motor is an age thing, I think...   once you have enough projects under your belt, you learn to "leave well enough alone" because the time saved means you can do even more crazy projects!   

Haha, perhaps! I'm not really a spring chicken either, just a dumb mechanical engineer who finds voltage and current more interesting than force and stress these days...

[james_s]

...Chinese crap on the shelves today. That's not an upgrade, it's a downgrade, with an added hack to make it work.

Ouch!

Earlier up the page I mentioned that I have several of these heaters to refurbish, so swapping the motor is a universal solution I can apply across all of the units. The perfboard test circuit may appear as a hack, but it's not representative of the final design.

I used to live in China. I saw all sorts of hokey products, especially in their domestic market. I understand where you're coming from by suggesting the reuse of the potentially "higher quality" original parts, but in the case of the fan motor, it's not something I'm interested in doing. Besides, this is just a fun side project; if my design choices allow me to explore some new electronics concepts, then all the better... right?

But why? What testing have you done to determine that the original motor is defective or deficient in any way? What knowledge are you basing that determination on? To me this looks like the attitude of "it's old so it's obviously no good" which is one of my pet peeves. I was not intending to sound like I'm bashing China, but I've watched so many products get cost engineered, becoming a little thinner, a little lighter, a little flimsier each year. A new shaded pole motor is gonna have the smallest amount of the lowest quality iron they can get away with, the smallest number of turns of the finest gauge of wire that will do the job, the thinnest bearings with the least amount of lubrication that will last through the warranty, they'll have shaved off every fraction of a penny they possibly can. It drives me nuts to see people replacing perfectly good original vintage parts for no good reason, it's wasteful and in most cases degrades the product. Something survives decades in original condition and then somebody throws it out just because it's old and sticks some new lower quality part in there. Granted this heater is not a rare and/or high value antique but the attitude is pervasive and ends up destroying a lot of nice vintage gear. It's only original once, and once and every modification is some part of the history lost forever. I would bet that there is absolutely nothing wrong with the motor in any of these, blow the dust off and put a few drops of oil on the bearings and they'll run for decades more.


As far as safety, these heaters have a completely exposed live element behind a grill with gigantic openings and no sensors to shut it off if it gets tipped over or blocked. It's fundamentally not a particularly safe design to begin with and there is not really anything you can reasonably do about that *however* if these characteristics are accounted for it is not particularly risky to use. You simply have to rely on yourself to operate it safely, keep it away from children, pets and flammables, don't leave it running unattended and don't leave it around for any random idiot to plug in and use. Used sensibly it's not going to hurt anyone, but it was built before the current culture of people expecting everything to be idiot proof and protect them from themselves so they don't have to.

[iroc86]

I hear you, james_s. I appreciate the discourse and validate your opinion on the matter. In general, I agree with you--I am positively adamant about recognizing the value of old stuff, be it the quality, scarcity, history, or just some intangible feeling. I have books on the subject I can share. I'm not being facetious, just recognizing that this is a really interesting topic with many cultural undertones. I've saved more tools, hardware, and old gear than I care to admit, often for no reason other than exactly what you pointed out: some part of history is lost forever. Hence this thread!

If material preservation truly didn't matter, why bother sinking time, money, and effort into refurbishing some old consumer appliance that would have likely been tossed in the trash anyway? If heating a space was the only goal, it'd be far more sensible to buy a brand new TPI/Markel heater (made in China) or some twenty dollar special at Walmart.

I agree with you about the safety aspects of this appliance's design. Any safety improvements I'm planning are more on the electrical side and less on the use cases.

As for the "why," it's because I want to. As you pointed out, this old Markel heater isn't particularly rare or valuable, so I'm not shy about changing out a few components. If I really need to defend my engineering judgement on the motor, it goes like this: 1) Bad bearings; not interested in sourcing and replacing them. Oil is a temporary fix. 2) Wires look a bit sketchy. Exposed copper in some areas. Paper insulation. Not dealing with that. 3) Windings feel brittle. Enamel flaking off. Hope that's not happening in the coil. 4) High temperatures inside case. Fifty years old. Thermal cycling, effect on winding, see previous. 5) How about a new motor? Triac speed controller, haven't done that before. Get to make another PCB and learn something new. COTS part. Check, check, and check.

There's preserving something just because you can call it original, and there's refurbishing something to make it functional again. By my requirements, "functionality" is also piece of mind.

[iroc86]

This is just a quick update with a few more simulation comparisons to the real world. I've finalized the fan speed controller circuit with an RC snubber and input filtering to better control dV/dt for the triac. The rise time has increased from about 300 ns in the previous version to almost 10 µs. dV/dt is 30 V/µs or less across the range of output waveforms, which is well within the operating characteristics of the triac. The dampening effect is necessary to avoid avoid spurious triggering due to the voltage/current phase shift from the inductive load. (The LC filter also tends to help here, taking the edge off of the rising waveform for RFI suppression.)

I also added a second RC phase shift network to reduce the control hysteresis associated with triac chopper circuits. The diac is triggered by C3 via C2, helping to isolate the control circuit when the diac fires. This feature isn't really necessary for a fixed fan speed, but I wanted some better adjustability when setting the speed during testing and evaluation. This so-called "gate slaving" technique is commonly used in lamp dimmer circuits. There are a ton of reference designs available online and I found a similar circuit in a commercial Lutron light dimmer switch.

All in all, I'm satisfied that the simulations can be duplicated fairly accurately on the bench. Next step is a new system schematic and rewiring plan.