Electrical Questions - RPC Controls

[J_R_Thiele]

Some pictures may be helpful.

Normally you would have run capacitors between L1 and L3, and L2 and L3, with L3 being the generated leg.

You are saying there are only two wires coming out of the idler and into the capacitor box. That would mean that ALL the run caps are on the same leg, That would explain why the voltage on the one leg is so high...

How many caps do you have in total and what are they rated at?

How are they being connected to those two wires?

 

[dalmatiangirl61]

Dug into the crimp connectors last night, had to use a dremel cutoff wheel to get thru the heat shrink, was kind of surprised to see corrosion here, my memory does not remember these being corroded, but it also remembers them being silver colored so....... My initial thought was mouse pee, but there are no other signs of mice in here, and there is plenty of cardboard for them to shred if they were in here. No corrosion going on with capacitors so I'm stumped.

At least one side of each of the crimps was not as tight as I would like, kind of surprised I did not replace these before. Not seeing anything else wrong so I'm going to put it back together with new screw tight splice connectors and test again.

There are 10 caps total, all 50uf 370vac rated. CSC is Commonwealth Sprague Capacitor, I remember Sprague made in USA caps being good, but don't know about the made in mexico variety. There are 4 strings of caps, 2 each of what is drawn below, one leg of each string goes to #2 wire, other to #3 wire (4 wires into each crimp, not 3 as mentioned in previous post).

Electrolytic caps can gain capacitance with age, but motor run caps lose capacitance with age.

 

[J_R_Thiele]

A 3 circuit terminal block as in the picture below makes it easy to add or remove caps when tuning and know you have a good connection.

The one in my picture is larger than needed, but the price was right.

From what you have written you have all 10 caps between L2 and L3.

For that number of caps I would expect to see 4 between L1 and L3, and 6 between L2 and L3.

How many HP is your idler?

 

[dalmatiangirl61]

A 3 circuit terminal block as in the picture below makes it easy to add or remove caps when tuning and know you have a good connection.

The one in my picture is larger than needed, but the price was right.

From what you have written you have all 10 caps between L2 and L3.

For that number of caps I would expect to see 4 between L1 and L3, and 6 between L2 and L3.

How many HP is your idler?

Its a Phase A Matic R20 , factory/custom/limited production (I'm not sure which applies best) built unit. Purchased used (cheap) and it did not run, so I replaced the caps, that was approx 10 years ago. Although I re-wired caps (new wire with spade connectors), I copied what was there already, so it is still in factory configuration.. My understanding is the idler was custom wound by Baldor, so its not quite a standard motor, beyond "a couple extra wraps on 3rd leg", I don't have any more info. No switches, timers, contactors, just a box mounted on top of idler with 10 caps and 2 wires marked 2 and 3 coming out of the top of idler.

 

[J_R_Thiele]

OK, I understand the situation now.

When will you be able to check the voltages when hooked up to machinery?

 

[dalmatiangirl61]

RPC is back together, don't think much changed, slight difference in readings, but probably just in the line voltage.
To neutral
L1 115v
L2 114v
L3 239v

L1-L2 232v
L1-L3 250v
L2-L3 276v

I'll get wired to a machine and report back in a few days.

 

[JST]

.............

L1-L2 232v
L1-L3 250v
L2-L3 276v

I'll get wired to a machine and report back in a few days.

That's some high voltage there...... But if it works it works.

 

[dalmatiangirl61]

Got the first machine wired in this past week, voltage readings are about the same with a load on the rpc, and after running for about a minute the thermal overload on the machine trips out.

I replaced the crimped copper tubes connecting the caps to rpc with a screw type butt connector, my plan now is to add some longer leads to those where I can access to add/remove caps while testing to figure out how it affects output voltage. And Sparky is going to come over next weekend with a capacitor tester to verify the values of the caps.

My thoughts, I purchased this used, maybe someone was in here before me and did something wrong, and I have just copied their mistake? If voltage is too high that means there is too much capacitance, but when motor capacitors go bad they loose capacitance, maybe some 100uf capacitors rolled across factory floor and got tossed into the 50uf hopper and labeled as such?

Feel free to add your thoughts.

 

[neilho]

If it were me (I know, it ain't ) I'd remove the caps from the circuit and start over, adding them back in one by one till the voltages were reasonable. Keeping track of changes on paper helps me a lot - too many variables for me to do it in my head.

Balancing current then becomes easier. Possible to check current at the machine while running, before the thermal trips? That'll give some hints as to where to add /subtract caps.

O'scope makes it all easier, (I gotta get me one someday) but getting the capacitance close enough is very doable by trial and error.

 

[JST]

Or, just use a boost transformer to get the voltage up to where it needs to be. That's more stable, and has no issues with no load voltage.

 

[dalmatiangirl61]

Or, just use a boost transformer to get the voltage up to where it needs to be. That's more stable, and has no issues with no load voltage.

Voltage is too high, not to low.

 

[CITIZEN F16]

That is the highest wild leg to ground I have ever seen,WTF! I agree with starting over, something is odd.

 

[J_R_Thiele]

The voltage is too high because there is too much capacitance in the system with the capacitors you currently have connected

The boost transformer would replace the capacitors as a way to boost the voltage in the generated leg. The boost transformer method works better at keeping an increased generated leg voltage under load. Its big disadvantage is that boost transformers are harder to find, and expensive to ship. Capacitors are much more available and affordable- so get used much more often.

I agree with neilho. Disconnect all the capacitors, start the RPC and record the voltages. Start adding capacitors one at a time between L1 and L3, and L2 and L3, and record the voltages between L1-L2, L1-L3, and L2-L3. The L1-L2 should stay pretty much the same, as it is what the power company supplies. Keep adding capacitors until L1-L3 and/or L2-L3 are about 5% above L1-L2. You reported earlier L1-L2 was 232 V, do 5% higher would be 243.6 V.

Now run the machine, and record L1-L2, L1-L3, and L2-L3. The generated leg voltages will drop. Hopefully they are within 5% of L1-L2. If the mains voltage is 232 V you want the generated legs to be 220 volt --or higher--. If you put a load on the machine the generated leg voltage will drop further, so you may want to add capacitors to get to the --or higher-- (plus 5%, not minus 5% of mains voltage)

Because you plan on running multiple machines you may want to set up the capacitors in the RPC to run the lower HP machines, and on any higher HP machines add a box with additional capacitors that only connect to the system when that machine is run.

 

[jim rozen]

T... Disconnect all the capacitors, ....

Again, start with this. Don't bother checking the line-to-neutral voltages, they're meaniningless in a system like this.

 

[dalmatiangirl61]

Called tech support again, got tech#2, where tech#1 instantly said voltage was too high, tech#2 looked up files and said voltages I am getting are correct. They have a solution, its a transformer, not exactly inexpensive. Pic below is how they suggest wiring it in, anyone want to hazard a guess at what type of transformer this is? Rating? I have a pile of transformers to look thru.

 

[JST]

Surely a buck-boost. Looks like he wants you to leave the caps in, and then put the transformer in place, wired for bucking down the voltage.

Rating for the current you need on the output.

I'm not crazy nuts about the idea and connection suggestion.....

 

[dalmatiangirl61]

Surely a buck-boost. Looks like he wants you to leave the caps in, and then put the transformer in place, wired for bucking down the voltage.

Rating for the current you need on the output.

I'm not crazy nuts about the idea and connection suggestion.....

I'm not qualified to judge either! Yes, caps left in place. Its an option they offer, they call it a voltage stabilizer, info is on the website. Cost wise I'm wondering if maybe that money would be better spent on a different unit. Pretty sure I still have a couple 20hp motors back in Tx, but not worth the drive, or shipping expense, so I can't just build from scratch. I'll go poke thru the transformers.

Edit: Output amperage of RPC is 54 amps, looking at transformers online it seems most are rated in VA or KVA, so not sure how to convert that to amps. And my search for "buck boost transformer" is showing what appear to be regular 120 to 240 or 240 to 480 transformers, and I don't think that is what I need. Any chance I can get a link to the buck/boost type I need so I can know what to look for in the transformer stash?

 

[dalmatiangirl61]

After doing some reading here I found a thread with basically the same problem Buck Boost Transformer Question

So doing some extrapolation.... The change in voltage is 40 volts multiplied by the circuit amperage, 54 amps, multiplied by 1.2 safety factor equals 2592 va, fudge factor it and call it 2.5Kva. Does this sound correct?

Further reading shows some people recommending the standard dry type transformer potted in a steel box, and others recommending 120/240 to 12/24/36/48vac transformers, I am failing to understand how these are interchangeable.

After searching here for a dry type that I thought I had, I called back to Tx, its sitting under the workbench there, f'me.

Searching ebay, will this work? GE 2.5 KVA Transformer, # A80L-0001-0273, Used, WARRANTY | eBay

Edit: Looking further online it looks like AR 20hp rotary converters start at $1,499....but I don't want the base model if its going to have same issues.

 

[J_R_Thiele]

For what its worth- here are my thoughts.

You have a used RPC from a known manufacture. You have replaced the original capacitors and believe the wiring is as designed.

Have you confirmed with the factory that the current wiring is correct?

The manufacturer reports their idlers are made by Baldor and are a specific design which is self starting, and a "true phase converter" and not a modified electrical motor. I do not know how to evaluate these claims. I have a 1 hp 3 phase motor that self starts. How it was modified to do so I do not know.

This forum has participants with a wide variety of backgrounds and experience. Most of our experience- and mine- is with RPC's made with conventional 3 phase motors. With those, I know what voltages to expect, how they are generated and how to modify them.

With this RPC I think you have a choice to make about where you get assistance. Is it to be the manufacturer or the forum.

The manufacturer -should- be the better choice. But---the different answers they provide are concerning- as is the need for a "solution" so the unit functions as it should. They will probably not provide any specific answers about the design of the motor.

If the forum is to provide useful answers be prepared for a lot of specific questions and requests.

I will start with asking if L1, L2 and L3 connect at the motor with 3 leads- or does the wiring visible resemble a motor with 6, 9 or 12 leads?

 

[JST]

part of what I don;t like is the connection of the transformer between the generated leg and another leg.

The better connection is likely to be from the generated leg to the source neutral, which is stable, and solidly connected to the "pass-through" legs, since you are starting with normal 240 single phase US voltage (120V-0-120V).

That voltage from neutral is not 240, and it is not 120, it is about 208V, or 0.86 x 240V. So you can use a 240V buck-boost, but the boost will be 86% of normal. A nominal 24V boost will turn out to be almost 21V. Nominal 12 V will be about 10V.

So, you connect it up for 240V, as a boost, connecting the low end of the winding to the neutral of the source, and the junction of the 240V and the boost winding to the generated leg. That gives you a fixed boost of about 21V, which will never be higher, even at no load. It's predictable.

If that is too much boost, connect the boost for 12V, which will get just over 10V with the lower input voltage.

With 240V in, the generated leg will start out 10V (21V) higher than the generated leg, and so it can fall 10V (21V) before dropping below the pass-through legs.

The generated leg is normally close to, but less than, the voltage applied to the motor (the pass-through voltage), because it is the back EMF of the motor, which must be less than the applied voltage, or no current can flow. that "less than" is the reason for the typical capacitors that provide a (less stable) boost in voltage.