Variable load device

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am looking for circuit or part that will actively route a high energy sine wave in the following manner:

Device/Part has constant input power wave A
Device/Part output Branch A1 receives main power on a variable load
Device/Part output Branch A2 receives A-A1 residual power

What should I be looking at?
 
Solid state amps or tube?


Let me try a guess. You want to take an amp putting out say 100 watts and run it through a thing that lets you send some amount to a dummy load and the rest to a speaker? You'd need something related to an L-pad maybe. SHows the amp a constant impedance but output is variable.

There you go, two L-pads on the same shaft, one wired reverse of the other. One for each load.
 
Thanks all for your input but it looks like DUG has understood my need the best. The electrical function required was described in the first post and it didn't specify an amplifier but hinted at line/load conditioning. The device/part needs to immediately route low frequency AC power on demand to one main branch output A1 while allowing the remaining power to be used on a subjugated residual branch output A2 with minimal loss ideally represented as A=A1+A2 and A2=A-A1 where A1 is described as a variable load.
 
Last edited:
Not sure what you mean by 'constant input power' - whatever is driving this automatically adjusts its voltage to deliver the same power whatever the load? Do you really mean constant input voltage?

If you want a useful answer then you will have to give us more information. At present we don't understand the question, and some of us may suspect that you don't understand the question either.
 
DF96
Yes, the wave pulse is constant and ideally should be used completely and no, I don't believe it will work automatically because the main branch will use from 1-95% of the total power output. The output must be differential because one branch is critical and variable and the other is not. The design MUST be able to make this power allocation in real time (less than 50 milliseconds) or it will not work properly. What leads you to conclude that I don't know what I mean?

mickeymoose
If what you mean is some kind of torque distributor, then yes, I need the main branch to use only as much of the power as it needs and the residual branch to take up what's left without competing with the main branch. The voltage will be constant so all that is left to do is insure that none of the current goes untapped.
 
Last edited:
gmaslin said:
What leads you to conclude that I don't know what I mean?
Statements like this:
Yes, the wave pulse is constant and ideally should be used completely
You appear to be mixing up power and voltage, and may in addition be trying to treat a low frequency problem as though it is an RF problem.

As I said, we need more information. What exactly are you trying to achieve?
 
Last edited:
DF96
Voltage is factor of power, where did you see me confuse the two? As I stated earlier, I have an AC power source that is fixed at a given voltage. I need a way to branch it so that one output has demand control and the other doesn't. My objective seems plain to me, what are you having trouble understanding? In addition, we are in the ELF realm so why bring up RF?
 
gmaslin said:
As I stated earlier, I have an AC power source that is fixed at a given voltage.
Where did you state that you have fixed AC voltage? Post 1 said "constant input power"; I queried that in post 6. Post 8 said "the wave pulse is constant", which did not answer my question because you did not say what parameter is constant.

I'm pleased that your requirements are gradually emerging. If I can now express what I think you are asking, but in conventional engineering terms:
You have a fixed AC voltage source, which is capable of supplying a particular current. (post 10)
You have a variable (resistive?) load A1. (post 1)
You want an additional (automatically controlled) load A2, to take up current not supplied to A1 so that the total current drawn stays constant. (post 1)
You say this uses "low frequency" power, which I take to mean 50 or 60Hz. (post 5)
You want the automatic circuit to respond within 50ms. (post 8)

Requiring a response in 2-3 cycles is quite demanding. You will need to monitor the instantaneous current and compare with a sine wave model, adjusting the controlled load via a servo circuit to maintain the total current correct throughout the AC cycle.

PS as this thread does not appear to relate to component parts for an audio system perhaps it should be moved to the Everything Else forum.
 
Last edited:
DF96
You have fairly surmised my intentions in 'conventional engineering terms'. I expect a flood of participation now that you have stopped me from confusing everyone. I hesitated to state the voltage was fixed because I may need step downs but the voltage will indeed be fixed within the operating parameters. I may be able to workaround a longer response time. What is the worst case response time of your approach? For the record, I disclosed the voltage was fixed in my reply to mickeymoose and stated this wasn't for an amp but never said it wasn't audio system related. I placed this thread hoping there was a part which performs the function I described but if there isn't, I have no objection to the mods moving it.
 
Last edited:
gmaslin said:
For the record, I disclosed the voltage was fixed in my reply to mickeymoose
OK, I can see that now in post 8.

The response time would depend on how clever the algorithms are, how much noise is present, and exactly how you adjust the controlled load. Fixed increments (switched by relays or triacs) or a big fat rheostat with a servo mechanism to adjust it? Remember that you can't move big things quickly, if you want them to last a while.
 
DF96
I'm glad we are beginning to understand each other. Noise on the source A is negligible if that is what you mean. I'm assuming that you would approach this with a logical gate controlling a heavy duty rheostat. What would you use to sense the current with this configuration? I didn't find a 20kW rheostat but if I do, what percentage power loss should I expect through this type of servo? Which of the two roads offered will be more efficient and which more economical?
 
Last edited:
Although I used to work in the power industry I am not a power engineer.

My guess is that you would sense the current using a current transformer and a DAC. Then you need a microcontroller to run the control loops. The power used in the controls will be small compared with the power you are wasting anyway.

This is not really a DIY job. You need to find a good power engineer, and a good control engineer with software skills
 
Sorry, I thought I was expressing an opinion not handing down a decision in binding arbitration.

Everything looks simple to people who don't understand the snags.

I know enough to know that this is a non-trivial, potentially hazardous, project. But what do I know; I only have a PhD in electronic engineering and about 20 years of industrial IT and SCADA experience.

You may get 'better' advice from people with less experience.
 
DF96
Did you not note the smiley after that comment? Lighten up, I wasn't casting any shadows on you or your acumen. Can you say the same? If I thought your opinion didn't have any weight, I wouldn't have made the effort to explain myself to you but keep in mind that just because you have pragmatic experience on how things have been done doesn't mean you can't be sparked into a fresh plane of thinking. I know the power I am talking about is not trivial but the solutions I have so far seen for this problem are clunky and impractical. There should be a way to create circuit leverage to get this done. If not, a passive solution where the subjugated branch has resistance/impedence that is a multiple of the maximum combined amp load of the main branch as you initially suggested might be the way to go. If this were the case, branch A2 would only draw comparatively to the main branch but this has serious efficiency limitations.
 
So, It looks like you need the function of a low frequency circulator.

Question: Is your power source a voltage source? i.e. low source impedance.

Next question: What voltage and currents?

I have some ideas that may result in constant loading ("instantaneous") but don't want to just throw them out there without correct starting information.

I guess one question no one asked is why you want your source to be always 100% loaded despite actual power drain from the source.

IMO: note to discussion of proper thread placement: probably should have been place in The Lounge or Everything Else.

However, it is here until a moderator (or one of those guys) moves it. So lets deal with it.

Need operating parameters.

I like a challenge.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.