March 20, 2013, Email of the Day, more ST gen head head scratching


A recommended utterpower modification

A recommended utterpower modification

The simple ST Head

The simple ST Head








Subject: Contact Form: ST Gen/ Light Flicker

From: David J.

ST Gen/ Light Flicker


I have a ST gen head from Tom at CGG and the harmonic winding wasn’t enough to maintain excitation voltage on the machine. Tom sent me a 120VAC AVR and instructed me to use one half of the machine output to provide excitation for the unit. The problem is that the 120VAC leg that is not providing the excitation is rock solid at 62 Hz no load and the one feeding the AVR is all over the place and causing the lighting on that half to flicker or pulse. Can you suggest a different way or a solution to provide excitation for my ST gen head? I am running the gen head with a Yanmar @2800 rpm and using a Fluke meter to measure the voltage and frequency for each 120VAC leg. Thanks, David J.

 George’s Answer:


This is an excellent question, and it can lead to a lot of learning.

We really need analyze the windings, and what they are doing, and since our rotating machine is designed to run at 60HZ, that’s where I choose to take measurements after we do the basic checks.

I believe in Kiss, and I do think it’s best to evaluate what you have in steps. On new heads we always document these readings and keep them, or we risk paying the price later. This way you know what you had when it worked right,  and you can do a compare later if you do have trouble.

Step one, a good meter is a must, if you don’t have a good meter, borrow one. (a Fluke is a good meter).

Read and record the resistance value of all windings.  Harmonic, Stators, and Field.

  • Consult your drawings and specs, are they in spec?
  • Don’t do anything to risk your warranty, check with your vendor before you conduct any further tests.
  • Most STs are designed to function with a standard rectifier, consider checking with your vendor, the radio Shack 30 amp rectifier they’ve had in stock for 30 years is normally plenty, the PIV rating need be well above the voltage peaks.

With the rectifier AC side wired to the harmonic, and the DC side to the field, apply 12 volts from a battery across the DC side (flash the field) with the same polarity as marked on the DC side of the rectifier..  Check all brushes and wiring,  (this is very important, if you are a regular visitor you’d note that poor connections in brush wires can cause weird behavior).

If the harmonic winding and the field winding are NOT working as specified on the name plate, (voltage and current) at 60HZ, then you likely have a problem with the Gen Head. Do you need to know more?  Is it in your best interest to apply a Band-Aid to mask the problem?

As for using one of  the stator windings (120VAC) to power the ST field, I Haven’t seen that, so I’d greatly appreciate others commenting if they have that setup, and they are happy with it?  I know that the people I have talked to think Tom works hard to make things right with customers, so I am not critical of his approach to help solve your problem.

If you are on grid you might try using another source of 120vac to drive this AVR just to see how it works… and if you have a link to drawings or other info, sharing with other DIYers here would be great.

The flicker issue: Since I don’t know about the circuit, hard for me to comment. My approach would be to put it on the other stator winding and  see how it works over there.

I would think if one stator is flicker free with a load, then the field itself is not the source of the problem, that would suggest a need to examine what’s going on current and voltage wise in the one stator that’s being use to drive the field, and most would want to look at both current and voltage in that stator, and display the output on a scope. There are very inexpensive scopes you can use with a lap top, a data analyzer would be great.

Others, keep me honest here.. I have little use for the Chinese AVR, I’d rather live with an acceptable voltage droop for most of what I do, as making power is far more important than keeping the voltage tighter than I require. Adding parts and complexity is fine if it meets your needs, but IF you are wanting a generator to make power in an outage, the kiss back up is a very nice option, no juice? Fall back to KISS designs..

As always, I suggest that people check every connection on the STs, and I personally eliminate the dog house and most of the crap supplied. The better connection is a properly sized twist nut, with tape over that, the volt meter needs to be off the set if you are wiring it up stationary. Big singles punish meter movements over time.

We all need Identify our goals when we build our generators. High on my priority list is a very quick fix IF I have trouble.

If you think the AVR is a must, then it’s best that you write down exactly why, and we do need know that AVRs can be very simple devices, and some quite complex and costing more than many generators. One thing we want to be sure of, and that is that the AVR is designed specifically for your gen head, OR it has an adjustment to limit the current in the field to assure it’s not over driven and damaged. And if it does have that adjustment, that you set it.. otherwise, consider a fuse in the field supply to assure it’s not damaged.

A great place to learn about simple AVRs is in Electrical repair shops where people contract to take care of assets from large to small for a construction company for instance. They know of the complaint in small portable generators.. “The guy said it was working fine till it ran out of fuel.”  In this case, a bare bones AVR can attempt to boost current as the generator slows down to the point where something in the field circuit fails, a marginal brush, poor connection, or it might just smoke the field if the AVR and field are not properly designed to dance together.






This entry was posted in DIYer Generator, Generators and tagged , . Bookmark the permalink.

9 Responses to March 20, 2013, Email of the Day, more ST gen head head scratching

  1. Dave S says:

    George; I really like the mounting of the diode bridge on the cover plate. Allows for quick check of z windings and flashing if field. It has also been recomended to mount the diode bridge in the brush bellhousing for air cooling. That install would make probing z winding, and flashing dangerous in my opinion. What say you?? Enough air flow under the block off cover?

    • George B. says:

      I fully expected this question, and I think it’s an example of personal taste. I do like the rectifier in the brush end bell housing best, and most ST’s have a lot of room there… and it’s easy to flash at the bell end.

      The Danger of mounting on the cover is removing same when the generator is running and shorting something to ground, and you are also disturbing the other connections when you remove the cover.. if the connections are really good ones, and the wire not brittle.. not to worry,

      If you mount the modern rectifier in the bell end, you need assure wiring is tied up!

  2. Ronmar says:

    One leg stable and one leg erattic almost sounds like a loose connection someplace. In theory, an AVR sampeling one 120V leg, will stabelize that leg. Since the other leg’s windings are intermixed, they are prettymuch along for the ride and should also be fairly stable. There of course is a caveat to this. The generator load MUST be balanced. If you are pulling a bunch of load off of the leg that is not being sampled, it will vary and the AVR will not correct for it. If the load is balanced, what is good excitation for one leg should be good for the other…
    IMO a better option is to connect an AVR across both legs and monitor the 240V combined output. At least you are looking at both legs at that point and changes to one leg will at least be partly corrected for. But the same caveat applies, the generator load MUST be balanced. The AVR looking at the combined output of both legs to achieve 240V. It dosn’t care how it gets it though, IE: 160V + 80V still = 240… If you do not have a need for 240, the best solution is to re-wire the head for 120V only and regulate to that with no balance issues. I would check with the AVR supplier, but you may be able to use that 120V AVR sampeling both legs by using a 240 to 120(2:1) transformer.

    Georg’s last advice is also VERY important to heed. There needs to be some appropriate protection in the line feeding the field from the AVR. The AVR dosn’t care what is going on, it just wants to maintain output voltage. If the gneerator is overloaded or the engine quits, as the generator head spools down and output drops, the AVR will continue to boost the field untill something goes up in smoke. A field fuse is usually used, sized just larger than the normal field current required for peak generator load/output. And of course, DO NOT shutdown with a load applied… More advanced AVR’s actually sample frequency, and cut excitation above or below a set freq/RPM to avoid this issue, but still use a properly sized field protection fuse to protect from overload.

    Good luck

    • George B. says:

      Ron you are ‘hands on’, always appreciate your input here..

      AVR’s are a study all by themselves, and we need make reference to our personal list of engineering priorities before we decide how or if we use them.

      One of the questions we might ask ourselves is how important is it that my generator make power when I want to run it? This gets into a lot of other issues from fuel quality and fuel storage, to excitation. Personally, I think of the harmonic winding and rectifier as the more simple solution, and whether we use it or not, it is the reason to own the ST; as we can use it when a more sophisticated means of excitation fails.

      We can add more complexity to the AVR by including that 2:1 transformer, and adding a DPDT switch for stator configuration, when stators wired for 120 VAC only, we drive AVR from there, when we select 120/240 VAC operation, we drive the AVR via the transformer just as you suggest. But at some point, we need ask… What are our loads and what quality of power do we need? If we are well inside of the ANSI standard for voltage at the distribution point from no load to full load, what does the average DIYer gain for the added complexity?

      The discussion of more critical loads and the needs for better quality power fascinates me. Most of the PC and computer equipment I see lives behind well engineered power supplies that can often run on crap power, and the equipment still well protected and performs well. Examples of where the AVR pays us regular folks dividends would be of high value here.

      Of course, we know there are as many views and ideas as there are professional and non professional engineers, and some engineers have their hats lined with tin foil, and in no way do I suggest you be one of them….


      Thanks again Ron!

  3. Butch(OH) says:

    Hi George,
    Being as I am an elctro idiot I cant comment on the fellow’s problem but as to your comment about the AVRs. I just may have lucked out and received the two finest ST heads to ever have left that empire but I have yet too see a need to any regulation beyond the harmonic for our usage. We must have our 220V for several reasons thus our 15KW (daytime) and 5KW (night time) sets are both so wired. With the engine governors set to give us 59-61 Hz average we are seeing voltages across the lines at 235-240 with the legs running 2 volts or less spread. We are not off grid of course but run the sets during the various outages we encounter and occasionally to make use of the waste oil we generate on the farm. Took the wife exactly 15 minutes to get used to watching the Kill-o-watt meter in the kitchen as she goes about her routines. Funny thing is she hears the engines take up the various loads and last time we had to run them she was washing clothes asked me Why is the engine “kicking in” more often than it used to when I run water?? I found the air pressure low on the water tank. Good wife eh? We have noticed that the ST head volts will drop 5 V or so as they heat up but have never seen voltages that would cause alarm as long as the Hz are in the range above. Going on to the next fellows question and wiring for 110 vs 220 operation I have not figured out why there is so much fuss about protecting each winding when wired for 220? It is simple to fuse each one separately as I have done with our units thus I am protected for both total amps and legs of each winding.
    All for now and enjoy the spring that is about upon us.

    • George B. says:

      your experience is typical WHEN you get a decent ST head.. quality varies..

      If you are an electro Idiot, perhaps you, me and a few professional engineers I know all are. We need define our requirements, and if they are met via KISS methods, they are still met.

      Hug your wife! I have one like that, she will even check her own oil, and listens for strange noises and tells me. I’m sure she’s saved me a million over the last 30 years..

  4. Ronmar says:

    You are spot on George, if you don’t need it, by all means KISS is by far the best way. But as you also mentioned, ST quality varies. I have come across a few who’s St heads won’t put out rated power at 60hz. Now it may be something as simple as bad wiring or bridge, and a very simplified solid state bridge like you have pictured, and going over their excitation circuit connections with a VERY critical eye(just replace all the terminal lugs:)), may find them the lost excitation current to boost their output to where it belongs. But I am thinking some of it is a bad wind or poorer quality copper used in the winding. At any rate, short of turning a low voltage ST into yard art, the AVR may be some users only option to achieve acceptable voltage output. I have experimented with them, but dont use one on my ST as it powers everything I have connected to it, even some finickey UPS’s, with the only addition to the harmonic field circuit being some capacitive filtering to clean up the field current into something more resembeling DC.

    • George B. says:


      And.. you just mentioned that critical load that many of us ARE interested in, or should be! In my opinion, we all need tip our hat towards that group of engineers that made the first decent Inverters with a decent sine wave output. Since much of this happened here in my home town of Kent, Washington, I choose to recognize them as the roots of what was good! But the result has been that these engineers split off and have formed a number of companies and made inverters that act differently. Some that are pretty particular as to the power quality they’ll latch onto as a source to power their internal battery chargers, OR to transfer the loads back to another source of AC.

      Some have even found their most popular Honda Generator did not have a high enough voltage at 60hz to satisfy the standard firmware in the inverter, and it ignored the Honda all together! I have seen transformers 1:1.1 ratio to bandaid this problem, but don’t we know we add inefficiency with it’s application!

      People in this community certainly should consider the inverter they buy, and know the generator they buy or build are compatible. Assuring that E is fairly high as per the ANSI standards >can< cut engine run times and battery charging times significantly. This is all data >any< engineer should consider BEFORE he starts to build something, whether he is a professional engineer, or a non-pro like many of us who build stuff. We do know that it will be hard for us to design things as poorly as some of what the pros have built! Thanks again Ron for adding so much value to this thread..

      • George B. says:

        I want to add one more bit of information for the engineer, again, whether you are a pro, a diyer, or best a pro+diyer. I will offer another load that can be harder to power off grid.

        A lot of us have pellet stoves, and we need consider how the engineers who designed these things designed it or were forced to design it. We also need realize that others who lord over them want to drive every nickle of added expense out of the design! Some were forced to reference the sign wave for timing.

        Let’s consider the fan for combustion, and the motor that runs the auger. What sets the rpm of that fan? In some cases, it’s the sign wave, and when you power that pellet stove at 55 HZ, the glass smokes up in minutes, and the stove runs like crap..

        It’s always best to know your loads and where you will put them. In the case of a not so smart pellet stove, leaving it on an inverter supply that always operates at 60.00 hz can be a good move.

Leave a Reply