ST Gen Head Growls, Email of the week

From: Chuck M.

Subject: ST Generator “Growling”
Message: Aloha George,
I have a belt driven ST 8 KW generator that changes from
a gentle hum to a distinct “growl” when the two 120V “legs”
get unbalanced. For example, both legs pulling about 1 KW each and my wife’s
iron kicks in, one leg goes to 2.43 KW and the growl starts. I have an AVR
sensing 240V (GAVR-15) and field current in the above case would increase from
about 2.5 amps to 3.1 amps. Growling also happened when I had a simple AVR (no
brand) sensing only one 120V leg. Have you observed this phenomenom or heard of
it before? Have any thoughts?
Chuck.. on one of those special events when Bill Rogers and I were both in the field together, we were in an off grid generator room where an ST head was making a funny noise.  Having been around a lot of 6/1_ST5 Gensets, I told Bill something is just not right here… but do note our host and friedn was not complaining as all was workign well according to his experience of a good many hours of running.  The set was supplying 120 volt loads only.  Bill got his hands on the volt meter and started taking a few measurements. He found the head was set up ‘center tapped’, with half the loads on either side. Bill read across the outside (the two hot ends) and did NOT find the 240VAC we were looking for. The off gridder had one stator coil hooked up backwards, and when we turned it around, the growling stopped.  As I remember, since the customer had no 240 volt loads, we decided to wire both stator coils in parallel and make things simple, and assure good load balance across the stator windings. But do note.. properly wired for 120/240 VAC operation, it should not growl even with an unbalanced load from my experience.
Your problem could be different, but check for this problem first.. I’dlove to see your follow up post here…
This entry was posted in DIYer Generator and tagged , , . Bookmark the permalink.

17 Responses to ST Gen Head Growls, Email of the week

  1. Chuck Moss says:

    Aloha George,

    Thanks for the reply. I verified that I’m getting 240V from the output leads. The generator is wired as center tapped, as I have some 230V loads in the house.

    In your post you suggested that I “check for this problem first”. Other than checking the output voltage is there anything else you think I should be checking?

    Incidentally, I loaded up the generator to a little over 4KW by turning on the kitchen oven and and the generator was making some more noise that when it was at 2KW and humming. The “some more noise” was similiar to the growl, but at about half the volume level.

    Next thought(s)?

    • George B. says:


      There are a lot of different brands out there now, and who knows all of what they’ve done to date..

      After you’ve ohmed out the stator windings, making sure they look the same, and used a megger to check the insulation, I’d check the field too..
      Next, I’d consider taking the AVR out fo the circuit and supplying my own DC from an external source..

      Does the head growl with a resistive load? We have a number fo guys familiar with Gen sets, they may have some ideas.

  2. Chuck Moss says:

    Aloha George,

    I ohmed the stator leads and didn’t see anything that would concern me. Configuration for ohming was to disconnect the output leads, neutral lead, and remove the strap normally connecting U5 to U6. The AVR is mounted in a wall mounted power distribution box and was therefore totally disconnected from the head. I left the doghouse voltmeter connected between U1 and U2. U1 to U5 ohmed at .9 to 1.0 ohm, as did U2 to U6. U1 to U2 ohmed at 1.8 ohms with the center tap strap connected.

    I don’t have access to a megger, but using my low voltage ohmeter all terminals ohmed off scale to ground and each other except through the intended stator leads.

    Please advise what you had in mind by checking the field.

    My head configuration is a ST-8, mfg by FuZhou E. N E&M Co, which I bought from Central Georgia Generator in mid 2009. Has been working fine when I test it on a monthly basis, other than occasion growling. It has roughly 100 hours of operating time. The harmonic windings are not used, and are capped off as the AVR generates the field current from the main output lines, using residual magnetism to get things going. The rectifier is simply sitting in the doghouse without anything connected to it, but there in case the AVR quits, I could simply hook up the harmonic leads & field wires to the rectifier and be back in (unregulated) business.

    As far as the loads go, in the 1 KW per leg more or less balanced, no growl condition the loads are daytime household loads for our house, and a sub-fed guest house. Three refrigerators, some lights, background always one clocks, TV’s etc. and a 1/2 HP pool pump running on 230V for probably 3 amps (.6 KW of the approx 2 KW total). Growl is initiated when a 1,440 watt iron would kick in and unbalance the load. In summary, mixed resistive & motor loads in balanced no growl condition, turning to a growl the moment the resistive iron load is added to one leg. Growl instantly stops when the iron kicks off.

    I’m open to your suggestions on how to take the AVR out of the circuit and providing my own source of DC. Nameplate shows field voltage of 61 V and 2.7 amps, although I measured voltage as 21 VDC on light load, and 35 VDC on a fairly heavy load (4-5 KW?). Field current at heavy load is a little over 3 amps. The same growling behavior was exhibited with the factory installed AVR that operated off the harmonic winding and sensed only one leg of the output.

    Hope I’ve been clear, and relevant. Glad to clear up any confusion if I can.


  3. George B. says:


    I have had few if any reports of growling.. I think we need let this topic perculate a bit, maybe some one will come alogn that had the same problem.

    As for the field, I just wanted yo to know it was good.. usually aroudn 17 ohms on the ones I am familiar with.

    Of course all who read this will ask.. what did your supplier have to say? Maybe others have the same growl?

  4. Thomas says:

    There are a number of threads about ST heads growling with unbalanced loads over on the lister engine forum –

    do a search for “growl” or “groan”.

  5. Chuck Moss says:

    Aloha George & Thomas,

    First, mahalo Thomas for pointing me over to the Lister forum. From reading the posts after a “Growl” search it is apparent that unbalanced load between the two household 120V legs frequently leads to growling. Some ST’s apparently growl at lesser levels (several hundred watts) and some tolerate a higher level of unbalance before growling. Several potential root causes identified, mostly relating to the magnetic fields inside, forces generated and movement of internal bits and pieces. It looks like main choices are tolerate, avoid/minimize unbalanced loading, and inserting a transformer between the generator and the household generator transfer switch/load. Transformer typically driven with 240V which will always have the two stator halves in balance. The two 120V household legs get driven by taps on the transformer, end taps are the two “hot” 120V leads, and the center tap becomes the neutral. At least that’s what I understand from reading the Lister forum postings.

    Needless to say the transformer solution has some trade-offs, 5-10% losses, cost, complexity, etc but appears to make using the whole power available on either 120V leg without growling or as much voltage drooping/boosting either leg. The biggest drawback looks like a few hundred dollars for a suitable transformer. Looks like load management/awareness is my way forward!

    George, field measured at 23 ohms. I haven’t talked with the supplier.

    In summary, looks like growling with unbalanced loading is a feature of the ST, and the transformer fix has its trade-offs.

    Still interested in other thoughts/ideas/experience/etc.


  6. George B. says:


    There’s a lot of variables in the ST design, some who claim to have built to the same Chinese Spec require far more field power than what we attempted to import. I measured an amazing difference in weight between one ST5 brand of head to another.. 40 pounds! So what did they leave out?? I mention this because a lot of people think STs are all the same.. they’re not, and some have less tendancy to growl. A friend mentioned there’s a chance the stators could even be wired differently.

    You mentoned an iron 1440 watt load? I wonder if that iron chops up the wave form in duty cycle fashion? You might tune an AM radio around and listen for the hash. Loose lams >could< sing like the Island birds with a chopped up load of that kind. I have one load on my outback inverter that makes it impossible to listen to the radio... even the one in my truck. It's the equivalant of a 556 timer running an SCR, after 10 minutes, it shuts off the bathroom light if it was left on. Loose lams, and other things could make an ST head growl more than the next one. A plastic fan is a sure sign of short cuts taken, the fan blades are the typical place to attach washer stacks for balance. Plastic fans often indicate that the effort to balance the rotating assembly was omitted as well. Some of the Mainland Chinese are making a stuff to sell, not necessairly to use, and they seem to be less worried about facing a firing squad when they export VS make local goods where they cheat their own people... or so I think.. Remember .. The ST is a fairly typical design, growling likely has more to do with how it was built than design... or so I think....

  7. Chuck Moss says:

    Aloha George,

    Mahalo for your thoughts/observations. I concur that how unbalanced forces/fields stimulate sound or not is likely more a matter of construction than design. One thread on the Lister forum had related level of growl to effectiveness of the varnishing of the windings. For whatever it is worth, my unit has a metal fan, and I haven’t come across any obvious problem that would flag poor quality. It came with a western style rectifier rated at 10A with 3.1 amp being the field current rating, and sealed bearings.

    On the iron it appears to be a bang/bang control law with maybe a 30 second cycle interval as it was being used. When it would cycle to “on” the growl would start and continue on a constant basis until the iron went off.


  8. bob g says:

    i am kind of late to the party here, but here is what i think i know about the problem,, and yes i have an st7.5 that will drive you out of your mind with an “air raid” siren sort of noise at times.

    these st heads we have to remember are primarily intended for 230-240vac operation in the far east, so each of the 4 poles on the stator are wired in series to get the full 230-240vac.

    now there came an opportunity for the chinese to sell to north american markets ( and others) that wanted not only 240vac but also 120 vac output.

    what the chinese did was to simply split the 4 coils between coil 2 and coil 3, using coil 1 and 2 for one leg of the needed 120vac, leaving coil 3 and 4 to provide for the other leg of 120vac.

    this is much different than first world heads where coil 1 and 3 would form one leg and coil 2 and 4 the other leg. providing 180 degree separation of the coils and therefore distributing the magnetics across the rotor from side to side.

    the chinese shortcut does not provide 180 degree separation, rather only a 90 degree separation, this has a tendency to load the rotor to one side as it is running, this makes for vibration which likely is aggravated/propagated by less than high quality brgs.

    what we have found is if you take down the genhead, separate the stator pole groups and reconnect them as coil 1 and 3, and coils 2 and 4, the magnetics are balanced equally across the rotor , even if the load on each leg is vastly different.

    we have reports of this being the only fix to this problem, even though new quality brgs go a long way toward quieting the issue, they don’t address the root cause.


    bob g

    • Chuck Moss says:

      Aloha Bob,

      Thanks for your interesting and informative input. I think I would be hesitant to rewire the internal guts as you have suggested, simply on concern for my skill level, not validity of your thoughts. For whatever its worth, I’ve ordered up a transformer via Ebay and will use it to provide 240/120V with the ST just seeing 240V loading and the “neutral” simply strapped together and not connected outside of the generator.

      From my understanding of your description this may not solve the problem completely, but I’m still expecting it to tone the growl down. In any case, it will allow me to do a better job of supplying unbalanced household loads.


      • George B. says:

        Hmmm.. great topic.. as the Student, let me ask…

        So…let’s assume we do have a single phase ST generator with stator windings NOT 180 degrees apart. What happens when we put these windings in parallel?
        What are the implications?

        • bob g says:

          when we either use all 4 poles/coils in series for 240vac, or… use all four pole/coils in parallel operation for 120vac, we end up with magnetic distribution evenly distributed around the stator core. at least as well as possible in a single phase stator.

          i would expect the growl to either go away completely, or be dramatically reduced by either the use of the transformer to step down the 240 vac to 120vac, or… by reconnecting the stator for 120vac operation alone, with both sets of coils connected in parallel.

          btw, i was not suggesting that anyone go into the stator and reconnect the coil groups, as it is probably a job for a motor rewind shop or someone that is very determined to do it himself and takes the requisite care to do it right.

          fwiw, ymmv and all that good stuff, 😉

  9. Chuck Moss says:

    Aloha Gents,

    My Ebay sourced 5 KVA Square D transformer, p/n 5S1F arrived yesterday after a slow boat trip from the mainland. Wired it up this morning to take 240V (only) from my ST 8 and put out 120/240V from the secondary. Works like a champ, and lo and behold, no growl from the generator under unbalanced loads from the two household 120V “legs.” Problem solved, as some $$$ expense, and at the expense of a some power loss, plus a few additonal wire interconnects, but all in all it lets the household loads be what they are without putting some apparent stress on the generator internals. Please note that although I have and 8 KW generator (vs 5 KVA transformer) my prime mover is only capable of driving 5 KW out of the generator.

    Mahao nui!

  10. George says:

    I am wondering if there could be a problem of sine curve Amp vs voltage sort of problem. I know that in certain applications of split phase motors that drive mechanical devices the transient electrically induced vibrations of the shaft compound upon the frequency of the driven device to produce a sound in the driven device sort of comparable to marbles rolling around inside of the device. Several years ago we found that by adding a 40 MFD capacitor across the windings of the motor we lagged the current out of phase with the voltage and voila …the marbles dissappeared. We figured that we had taken the current approximately 90 degrees out of phase with the voltage. I am wondering if in a generator the same sort of situuuation could be causing some of the growling.

  11. George says:

    Chuck Moss……..
    Sounds as though the xformer is acting a little like a smoothing choke.

  12. bob g says:

    the problem with st heads is this
    they are designed and wound to be 230volt 50 hz machines, and they change the turn count and speed them up for 240/120 volt 60 hz

    what they did not do is this

    they split the 4 poles as would be expected to enable 120 – 120 volts
    but they split it using two pole side by side and not opposing one another.

    it is very easy to simply split between pole 2 and 3 to separate for 120 operation, but if you load one leg you only load one side of the stator! so pole 1 and 2 are side by side and form one 120 leg, then 3 and 4 side by side form the other 120 leg. half of the machine is loaded while the other half is not, this causes the rotor to work toward one side of the case and makes more noise in many of these units.

    if they had split all 4 stator poles, and connected pole 1 and 3, and pole 2 and 4 the loading of each pair would be balanced across the stator 180 physical degree’s, not 90 physical degree’s

    i have a warning in some old generator text books warning about such interconnection causing an unbalanced magnetic loading of the stator, which will result in odd vibrations and noise.

    what we don’t know is how many are exported configured in this manner, but suspect the number is quite high as a percentage.

    some of these gen heads are horrible in the amount of noise they put out, my st7.5 i have on my unit sometimes will howl like an air raid siren, and can be heard for several blocks all around! ear protection is a must when it goes into its fits, i finally just made sure to do all my testing with carefully balanced loads on both legs.

    when it it put into service it will likely just be reconnected as a 120 only machine, which will force equal distribution across he stator.


    • bob g says:

      crap, i didn’t see we had this discussion over a year ago

      please delete my response if you feel it unnecessary George

      bob g

Leave a Reply

Your email address will not be published. Required fields are marked *