reg/rect Cooling

[Anonymous]

A well designed reg/rec (e.g. electrex) with robust components will be able to work flat out at max load in still air. The cooling fins are designed to have sufficient surface area such that the temperature difference between the air and the fins creates a stable temp in the reg/rec.
The heated air will not remain huddled round the reg/rec even while the bike is stationary. It will try to rise and will disperse due to convection and the principles of Brownian movement. At a given surface area, the reg/rec will not be able to get any hotter as the increase in temperature gradient will cause an increase in the rate of heat transfer.
The same principle ensures that aircooled engines don't just get hotter and hotter whilst in traffic. Eventually they stabilise. It may feel uncomfortable to you but if the manufacturers have done their sums right there will be no detrimental effect to the engine.
The main reason these things fail is probably more due to under-specification of components by the bike makers to save money. They fail when they get hot because the heat is an indication of how hard they are working.
Extra cooling can do no harm, heat is generally bad for electrical components. But if your reg/rec needs to be hung in the breeze to avoid premature failure, then it wasn't very well made in the first place.
I've had Electrex reg/recs on both of my GS's for some years now with no problems, and they're in the standard position hidden under the battery box (750) or behind the side panel (1000). They are much more heavily finned than the standard items and I would be willing to bet they use much higher grade componentry.

[lecroy]

"The heated air will not remain huddled round the reg/rec even while the bike is stationary. It will try to rise and will disperse due to convection and the principles of Brownian movement. At a given surface area, the reg/rec will not be able to get any hotter as the increase in temperature gradient will cause an increase in the rate of heat transfer. "

Wow, this is getting way over my head... On my old 83 bike, the fins are not mounted vertical, they are horizontal. So, when you calculate the effective surface area for convection cooling you are only looking at the tips of the fins for the most part. On that bike, while in motion, it makes sense that the air travels in a horizontal flow and why the fins are in this direction. Same holds true for the block. Now if I sit that same air cooled bike on a dyno and load it for very long, it's not going to like it. I call this the Being Stupid Principle (BSP). Will it (engine or electronics) fail if engine is left to idle for long periods with only convection cooling? Too many things to consider. Why run the bike while you blab on end to your buds? I guess that was my point.

"The main reason these things fail is probably more due to under specification of components by the bike makers to save money. "

I guess I just have not seen that many problems with any of my bikes to comment on this. The one time I did have a failure it was because when I was riding the battery opened. I got something like a load dump which took out every light bulb on the bike along with the electronic tach and regulator. Could the tiny guys have protected against this kind of failure, sure. I won't hold that one against them though.

I am not sure what you mean when you say "higher grade componentry". Assuming you are talking the temperature rating for the devices, it has nothing to do with the failure rate for a given device. As long as the designer takes the temperature drift into account, it is very possible to run 70 degree commercial grade parts at 105 and higher with the same reliablity as their automotive grade brothers.

I think when I looked at the schematics for that stock GS regulator that failed they had some kind of three phase triac schem that used the stator as a dump. This has been several years ago, so don't quote me on this one. But I think you could make a better charging unit using a small buck convertor that runs in current mode with a voltage limit what would be much more efficient than the stock part.

[Anonymous]

With regard to the cooling issue, you are assuming that the reg/rec has been designed to have some airflow over it. This may not be the case.
Undoubtedly the engine needs airflow at high load and the designers take this into account. If they had to assume zero airflow the fins would have to be huge, but it could still work. Because of its location, the designer would be reasonable to assume that an engine would receive good airflow and that there would be more as the load increases.
Engines get a lot hotter than reg/recs. The finning on reg/recs is enormous compared with engines, and can rely on conduction of heat to the air with zero(or very little) airflow.
You are right in saying that the design can affect the efficiency and therefore the amount of heat produced.
Its possible for there to be 2 components nominally rated for, say, 70 degrees, one of which is cheap crap and will fail sooner, despite what the maker may claim. You get what you pay for.

[lecroy]

With regard to the cooling issue, you are assuming that the reg/rec has been designed to have some airflow over it. This may not be the case.
Engines get a lot hotter than reg/recs. The finning on reg/recs is enormous compared with engines, and can rely on conduction of heat to the air with zero(or very little) airflow.
Its possible for there to be 2 components nominally rated for, say, 70 degrees, one of which is cheap crap and will fail sooner, despite what the maker may claim. You get what you pay for.

Oops, I didn't think I was assuming anything. Just pointing out effective surface area for a hoz. mounted fin.

Does an engine get hotter than the electronics? Where on the engine are we talking? We used to use the outlet coolant from the radiator and feed this into a small heat exchanger that cooled the electronics.

Would be interested to see some data on that last comment. Who do you feel makes better and worse of the same parts? Are you calling a part a whole assembly, or a discrete component like an SRC or IGBT? Plastics have made such a big change in electronics during the past 12 years that we hardly even use ceramic anymore. You you think the ceramic parts are better because they cost more?

On my old larger air-cooled GS engines there is no room for fins. No air flows between the cylinders! Of course they only run under load for a few seconds.

[Anonymous]

You're making me tired here.

Effective surface area? Look, bigger fins equals more cooling. If the fins on an engine were big enough you wouldn't need any airflow. Many electronic items in household and industrial goods are finned for cooling but have no airflow over them. I'm suggesting that a reg/rec may need minimal airflow to achieve a good rate of heat transfer.

Do engines get hotter than electronics? In general, yes. OK pedant go and find me an example of something to prove me wrong, but you know where I'm coming from.

On the last point, you also know the answer to that. You know full well that I'm not talking about expense for the sake of it or rarity value. I'm talking about a designer trying to meet a budget and choosing the cheaper option in the design and/or the specification of discrete components. I'm also talking about 25 years ago in Japan. There are many examples of this I could give you both from my experience with bikes and in my work, but I can't be bothered.

I'm just trying to help people on this website. You?

[lecroy]

"I'm just trying to help people on this website. You?"

I'm just trying to help you.


So, for example if you have a block of aluminum, say 2 x 2 x 2. Will this block cool the same as one that has cuts make into it to create the fins? If the fins are mounted vertical in air it will. If it is mounted horizontal it won't. If it's in a perfect vacuum it's the same. In no air flow the virgin block will provide the best heat dump.


"Do engines get hotter than electronics? In general, yes. OK pedant go and find me an example of something to prove me wrong, but you know where I'm coming from."

Yea, but I'm a detail kind of guy.

"You know full well that I'm not talking about expense for the sake of it or rarity value."

Sorry, but I didn't. That's why I asked you to clarify what you were saying.

I'm talking about a designer trying to meet a budget and choosing the cheaper option in the design and/or the specification of discrete components.

Yes I agree. I can design crap for the same amount of money as I design something good. But more often that not, not the other way around. Automotive electronics, at least here, is a tough business. Everything is cut to the penny. If you want to make your design more robust and it added money but was not part of the original spec. you won't get the business. Keep in mind also that warrenty can play a big part in a companies business. Where would the auto parts store go without failed parts? I guess they can continue to sell those fuzzy dice.

"I'm also talking about 25 years ago in Japan. "

Does not ring a bell with me.

"There are many examples of this I could give you both from my experience with bikes and in my work, but I can't be bothered."

Too bad. I would like to hear them. I just have not seen any big trends. If anything, I would have to humble myself and say they really did a lot to help the electronic industry in every way. Even the advances in no lead solder seem to come from Japan.

"You're making me tired here. "

Go take a nap!

[Anonymous]

We're not comparing a 2x2x2 block versus the same block cut for fins. In reality you would make the thing bigger by adding fins.

Also in reality there is rarely "zero" airflow. A reg/rec under a battery box is going to get a fair amount of air movement around it at most speeds above stationary. As you said, following Granny.

An example from my work is that recently we started to see failures on a particular PCB. The failure analysis found that the part vendor had changed component suppliers for some of the components on the board in an effort to maintain profitability in the face of the continued slump in the semiconductor industry, which has focussed attention on costs. They had found a supplier who could supply an identical (nominal)spec part at a much lower cost. Returning to the original supplier cured the problem.
This is not an isolated occurence, but is happening all the time, both on current lines and on the introduction of revised items which turn out to be less reliable than the old item due to the constant downward pressure on costs. Cheapest components, cheapest design and cheapest production methods.

My point is this: Its possible to put a good quality well designed RR in the same (less than ideal?) location and it be less likely to fail than the original which may have been compromised by production needs. Proof? My Gs's now have better charging systems than they did when they left Hamamatsu. More proof? Many bikes produced in Japan at the same time had charging systems which have been 100% reliable to this day.

Look lets quit all this academic debate. My fundamental points are correct as are all the points you are making. You are over analysing my statements.This doesn't contribute anything to the discussion, which is reg/rec cooling.

You are patently a qualified and knowlegeable person, so I would like to ask you this: Is it necessary to improve the cooling to a regulator and rectifier on a typical early GS?

Also: If so, is this still the case once the charging system has been modded by the fitment of an aftermarket reg/rec such as an Electrex?

My opinion?
In the case of the first question, don't know, I've never had a GS where the original reg and rec have lasted long enough.
In the case of the second question, I would say no as I have never had a problem with the replacement Electrex's I've fitted. My belief is that these RR's are robust enough to tolerate less then perfect cooling.

[redliner1973]

OK guys, here is a simple one......Underneath the airbox is the opening or the intake. In a four cyl engine, there is alot of air in that box.....whilst the engine is running, the seat, sidecovers, and airbox form a kind of 'chamber'. in this 'chamber' if you will, air is being drawn in from all the little 'nooks and crannies' in this area and going onto the airbox inlet. whislt this air is around, and moving through this area, it cools the regulator. hence the reason it is in the location it is. lots of airflow, and protected from the elements at the same time.......!

[lecroy]

"An example from my work is that recently we started to see failures on a particular PCB. The failure analysis found that the part vendor had changed component suppliers for some of the components on the board in an effort to maintain profitability in the face of the continued slump in the semiconductor industry, which has focussed attention on costs. They had found a supplier who could supply an identical (nominal)spec part at a much lower cost. "

Again, I would need much more detail. But the root problem may not have been that the part was changed but rather that the part you are talking about was not designed for worst case conditions. All devices will have tolerences on them. If the designer does not take all of them into account in his design you run into problems.


I though you could follow the 2x2x2 block and the idea of effective surface area. Just the basic idea that for the same outside dimension that you can lower the resistance by adding surface area. Motorola had a good book on the subject that you can get for free. Maybe it would help.


"Also in reality there is rarely "zero" airflow. A reg/rec under a battery box is going to get a fair amount of air movement around it at most speeds above stationary."

I believe this was my original point.


"Is it necessary to improve the cooling to a regulator and rectifier on a typical early GS? "

In my experience this is not the case. I have yet to see a heat related failure on any of my bikes over the past 30 years. I have purchased one of my 83s in 84 and the other around 87. Both have all stock electrical parts. The problem is that I tend to ride Suzuki and have owned a very limited number of bikes. Maybe 20 or so. So the sample just is not high enough to make any kind of a confidence call. I went to the Electrex website, but they do not seem to supply any data on required air flow, max. ambient temp. or mounting that I could find. There was also nothing on max. operating current, voltage. Worse, nothing on the regulation. I guess they could sell you anything and it would be within their spec.

[Anonymous]

reg/ rec

I put my reg/rec up front where the horns were had to make bracket ran wire under gas tank and spliced in 6 inches of wire so it would reach stator gets lots of AIR TO STAY COOL you can touch it and its warm not hot like it was before when it was under battery hope this will help you all out took me 2hours to move it. 1982 GS1100GKz

[Anonymous]

"I'm just trying to help people on this website. You?"

I'm just trying to help you.


So, for example if you have a block of aluminum, say 2 x 2 x 2. Will this block cool the same as one that has cuts make into it to create the fins? If the fins are mounted vertical in air it will. If it is mounted horizontal it won't. If it's in a perfect vacuum it's the same. In no air flow the virgin block will provide the best heat dump.


"Do engines get hotter than electronics? In general, yes. OK pedant go and find me an example of something to prove me wrong, but you know where I'm coming from."

Yea, but I'm a detail kind of guy.

"You know full well that I'm not talking about expense for the sake of it or rarity value."

Sorry, but I didn't. That's why I asked you to clarify what you were saying.

I'm talking about a designer trying to meet a budget and choosing the cheaper option in the design and/or the specification of discrete components.

Yes I agree. I can design crap for the same amount of money as I design something good. But more often that not, not the other way around. Automotive electronics, at least here, is a tough business. Everything is cut to the penny. If you want to make your design more robust and it added money but was not part of the original spec. you won't get the business. Keep in mind also that warrenty can play a big part in a companies business. Where would the auto parts store go without failed parts? I guess they can continue to sell those fuzzy dice.

"I'm also talking about 25 years ago in Japan. "

Does not ring a bell with me.

"There are many examples of this I could give you both from my experience with bikes and in my work, but I can't be bothered."

Too bad. I would like to hear them. I just have not seen any big trends. If anything, I would have to humble myself and say they really did a lot to help the electronic industry in every way. Even the advances in no lead solder seem to come from Japan.

"You're making me tired here. "

Go take a nap!

Sorry, had to jump in here.
"So, for example if you have a block of aluminum, say 2 x 2 x 2. Will this block cool the same as one that has cuts make into it to create the fins? If the fins are mounted vertical in air it will. If it is mounted horizontal it won't. If it's in a perfect vacuum it's the same. In no air flow the virgin block will provide the best heat dump."
Actually, in a vacuum, the finned block will still radiate more heat at a faster rate. There does not have to be air to radiate heat, thus the term "RADIATE" derived from radiation.

[lecroy]

"Sorry, had to jump in here. "

Hey, that's the idea. Why say your sorry?

"Actually, in a vacuum, the finned block will still radiate more heat at a faster rate. There does not have to be air to radiate heat, thus the term "RADIATE" derived from radiation."

Yep, you have me on that one. I have done some testing on electronic parts in a deep vacuum (2 x 10-7) and find that you really need conduction into the outside world to get any cooling. Of course I wasn't making my parts glow. Radiation provides little cooling. Convection also provides some cooling but there is not much stuff left in a vacuum this low. I was blabing about the convection kind of cooling in this case. The kind that matters on your bike. Somewhere on my dickey site there is a bunch of info on designing electronics for space flight. Fun stuff if your a G'ee'k.