Vfd application
- Thread starter tw/nci
- Start date
- Status
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
110611-0858 EDT
Independent of what modification you would need to do to the VFD to make it work, I believe I can correctly say the VFD would be less efficient than the simple SCR circuit.
Fundamentally the SCR is a single series diode drop, but somewhat higher than a plain diode. Whereas a VFD will have at a minimum two diode drops and also other power losses. There is much more circuitry in the series path from input to output in the VFD than in the SCR circuit.
.
Independent of what modification you would need to do to the VFD to make it work, I believe I can correctly say the VFD would be less efficient than the simple SCR circuit.
Fundamentally the SCR is a single series diode drop, but somewhat higher than a plain diode. Whereas a VFD will have at a minimum two diode drops and also other power losses. There is much more circuitry in the series path from input to output in the VFD than in the SCR circuit.
.
Last edited:
I guess it could but why would you?
A VFD has two stages of conversion. The AC input is rectified to DC and then converted to a variable frequency output.
A SCR regulator in the three phase supply is all you need at the power stage.
You might want to consider burst firing with zero crossover switching.
There are a number of suppliers who make such modules.
Caledon Controls is one.
kwired
Electron manager
- Location
- NE Nebraska
But typically the voltage is also varied with the frequency and this would change the heat output.
As others have mentioned there are other and better ways to regulate a resistive load.
Electric-Light
Senior Member
For something like that, voltage modulation using tap changers and an multi-tapped auto transformer is more efficient and this setup does not cause harmonics at all unlike phase controlled system.
A transformer with 335v and 415v taps will let you have 50, 75 and 100% outputs.
Motorized variac control provides stepless modulation, however it is far costlier.
A transformer with 335v and 415v taps will let you have 50, 75 and 100% outputs.
Motorized variac control provides stepless modulation, however it is far costlier.
Last edited:
ptonsparky
Tom
- Occupation
- EC - retired
Curious. Are solid state relays capable of switching fast enough for this type of application?
Electric-Light
Senior Member
What do you mean "fast enough"?
Some are simple plain jane triac. Some have zero crossing turn on circuit.
Former is fast enough to be used for phase control power control. It is also used for inductive loads. Latter is used for resistive and capacitive loads. Zero crossing turn on in contraindicated for major inductive loads such as multi kVA transformers and motors as it actually increases inrush current on inductive loads.
You could also design whole cycle based system that do not cause much harmonic distortion (i.e. 15 cycles on, 35 cycles off) but you need to stay out of control frequency region that is attributed to annoying flicker. I believe that is around 8-10Hz. Even a fractional percentage voltage fluctuation at that frequency will cause noticeable and annoying flicker on lighting.
Last edited:
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
110612-1440 EDT
ptonsparky:
For virtually all heater type applications speed of turn on and off is not a significant factor. The thermal time constant of the item being heated is very likely to be in the multi second or longer range. Thus, on-off modulation, as suggested by electric-light, with a zero-crossing SSR would work fine.
electric-light:
Explain why zero-crossing turn on creates an inrush problem with ferromagnetic inductors vs random turn on.
.
.
ptonsparky:
For virtually all heater type applications speed of turn on and off is not a significant factor. The thermal time constant of the item being heated is very likely to be in the multi second or longer range. Thus, on-off modulation, as suggested by electric-light, with a zero-crossing SSR would work fine.
electric-light:
Explain why zero-crossing turn on creates an inrush problem with ferromagnetic inductors vs random turn on.
.
.
Electric-Light
Senior Member
cadpoint
Senior Member
- Location
- Durham, NC
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
110612-1730 EDT
What happens is all dependent upon what is the residual flux. Generally a core will not be at a zero residual flux state. A zero voltage turn on point can be quite bad or very good. With a random flux state at turn off shifting turn on to a voltage peak will reduce the worst peak inrush current, but it does not eliminate inrush current.
If the residual flux state was known, then the correct phase angle from 0 to 360 could be picked and achieve no inrush current.
With an unknown flux state an approach is to apply a small volt-time integral to the input and gradually increase the magnitude to allow the core to redistribute its mean flux state. This can be accomplished without a phase shift controlled circuit by using a series resistor that is shorted after a time period. Somewhat automatically accomplished with a negative temperature coefficient thermistor. But in high power systems the thermistor needs to be shorted after time.
.
What happens is all dependent upon what is the residual flux. Generally a core will not be at a zero residual flux state. A zero voltage turn on point can be quite bad or very good. With a random flux state at turn off shifting turn on to a voltage peak will reduce the worst peak inrush current, but it does not eliminate inrush current.
If the residual flux state was known, then the correct phase angle from 0 to 360 could be picked and achieve no inrush current.
With an unknown flux state an approach is to apply a small volt-time integral to the input and gradually increase the magnitude to allow the core to redistribute its mean flux state. This can be accomplished without a phase shift controlled circuit by using a series resistor that is shorted after a time period. Somewhat automatically accomplished with a negative temperature coefficient thermistor. But in high power systems the thermistor needs to be shorted after time.
.
ptonsparky
Tom
- Occupation
- EC - retired
I use 480V SSRs to switch 480v controls with 3-10vDC. Finding small ice cube relays with contacts rated at 480v is difficult. I believe they are zero crossing and was just curiious if I could use them to control heaters via triac outputs on a plc. Not a likely project at the moment, just thinking.
Thanks,
Thanks,
gar
Senior Member
- Location
- Ann Arbor, Michigan
- Occupation
- EE
110613-1954 EDT
You can use an SSR to control a heater. Whether zero-crossing or not is probably of no importance. Less RFI with the zero-crossing.
A Triac output is not directly suitable for switching DC, but could be made to work with additional circuitry. A transistor (DC output) or relay contact is the best choice.
.
You can use an SSR to control a heater. Whether zero-crossing or not is probably of no importance. Less RFI with the zero-crossing.
A Triac output is not directly suitable for switching DC, but could be made to work with additional circuitry. A transistor (DC output) or relay contact is the best choice.
.
Electric-Light
Senior Member
You can, but be absolutely certain to install non resetting thermal protection. Triacs are notorious for failing closed following a surge, which means that the heater will stay on.
Relay contacts can weld too if it closes into a shorted load, but under normal conditions, its not nearly as common.
If you do not need the speed of modulating partially through each cycle like a light dimmer, it really doesn't matter if you use zero crossing or normal triac.
- Location
- San Francisco Bay Area, CA, USA
- Occupation
- Electrical Engineer
I for one am still curious as to why you were wanting to know.
weressl
Esteemed Member
The ASD would certainly have a problem with not 'seeing' and inductive component, but that the easy part of the issue. Why would you want to spend 10 times as much money, or more, when the currently available replacement SCR controllers will do the job?
- Status