Thank you very much for your fast reply.
It is very strange situation here. I have read the instructions one more time and performed a factory reset. I think it is much better now. There are still fluctuations in the voltage but in the range of 3-4 volts. It was nearly 11 volts before.
To answer the questions:
1. The input levels of the TDA board are steady at 23.9 volts DC (24volts recommended)
2. I had a transformer and made a linear power supply 12 volts (1A) for the signal generator and 24 volts (2A) for the power amplifier. I am using a classic schematic with regulators (7812 and 7824), caps for filtration and some drosels (coils). The power supply is steady when measuring under load and under no load.
3. Yes the output voltage fluctuates both when under load and when the turntable not in use. I was measuring directly on the output.
One more question:
I am not sure that I understand the "phase" thing. I left is exactly as the factory settings are. No adjustment there. Do I need to make something in this menu to make it work more stable? As I also have an Arduino tachometer I have noticed that with a controller attached the RPMs are not steady, unlikely when powered directly from the grid... does it means that I have also fluctuations in the frequency. Sadly I do not have a HZ meter and cannot check.
Thanks for your help very much.
It is very strange situation here. I have read the instructions one more time and performed a factory reset. I think it is much better now. There are still fluctuations in the voltage but in the range of 3-4 volts. It was nearly 11 volts before.
To answer the questions:
1. The input levels of the TDA board are steady at 23.9 volts DC (24volts recommended)
2. I had a transformer and made a linear power supply 12 volts (1A) for the signal generator and 24 volts (2A) for the power amplifier. I am using a classic schematic with regulators (7812 and 7824), caps for filtration and some drosels (coils). The power supply is steady when measuring under load and under no load.
3. Yes the output voltage fluctuates both when under load and when the turntable not in use. I was measuring directly on the output.
One more question:
I am not sure that I understand the "phase" thing. I left is exactly as the factory settings are. No adjustment there. Do I need to make something in this menu to make it work more stable? As I also have an Arduino tachometer I have noticed that with a controller attached the RPMs are not steady, unlikely when powered directly from the grid... does it means that I have also fluctuations in the frequency. Sadly I do not have a HZ meter and cannot check.
Thanks for your help very much.
Last edited:
I should have been more specific: By input levels to the amp, I meant the AC signal (sinewave) not the DC voltage.
If would help if you had an o-scope to look at the output as well as making sure there is no ripple on the DC line.
The factory settings for phase should be OK. I doubt the SG4 frequency is fluctuating. Do you have the eddy current brake disabled on the 401?. The 401 uses an AC induction motor which will be sensitive to voltage as well as frequency. Running at a lower voltage may affect speed stability.
If would help if you had an o-scope to look at the output as well as making sure there is no ripple on the DC line.
The factory settings for phase should be OK. I doubt the SG4 frequency is fluctuating. Do you have the eddy current brake disabled on the 401?. The 401 uses an AC induction motor which will be sensitive to voltage as well as frequency. Running at a lower voltage may affect speed stability.
I noticed this amp on sale at PartsExpress and it seems like a potential candidate for the back end of the SG4:
TPA3116D2 2x50W Class D Stereo Amplifier Board with Volume Control
It has a built-in attenuator, which should eliminate the need for a separate pot on the SG4 output. It's also d**m cheap.Has anyone played with this unit?
-b
As someone who's getting ready to start in on an SG4, I'm curious to know the answer to this question as well. (Ordering from PE is a plus for me right now, as I've had long delays for a few recent international shipments.)
Hi Orlin,
Are you using the recommended Amgis transformer? I have found that the 7492 amp has problems with lesser quality transformers. Mine works with a potted transformer but not with any of the other cheaper toroids.
I remember seeing a Bel 530-LP-12-1900 transformer recommended as well.
Regards,
kffern
The TPA3116 amps are fine as long as you're using an output transformer to use with 110 or 230V motors.
You cannot use them with BLDC motors unless you dig inside the motors and separate the windings as the amp outputs are not referenced to ground.
When using these amps it's often a requirement to use a resistor between the amp and the output transformer as the low impedance of the transformer can make the amp 'think' it's driving a short circuit and shut down. I normally get around this by using an NTC resistor, specifically NT 5D-9 which has a high resistance when cold, but drops to less than 0.25 ohm when hot.
You cannot use them with BLDC motors unless you dig inside the motors and separate the windings as the amp outputs are not referenced to ground.
When using these amps it's often a requirement to use a resistor between the amp and the output transformer as the low impedance of the transformer can make the amp 'think' it's driving a short circuit and shut down. I normally get around this by using an NTC resistor, specifically NT 5D-9 which has a high resistance when cold, but drops to less than 0.25 ohm when hot.
Measuring the winding resistance will be the most accurate way to determine the static load impedance (the impedance increases as the motor speed increases), but you can get reasonably close by computing it: R=16²/Power. The impedance doesn't fall below 8Ω unless the power is higher than 32W. It should be OK.
Even in bridge mode, you will need an output equal to 22.6VPP to generate 16VRMS. Using a 24VDC supply, it should be doable, but you may see some minor clipping at start up; running at a reduced voltage should be OK.
Even in bridge mode, you will need an output equal to 22.6VPP to generate 16VRMS. Using a 24VDC supply, it should be doable, but you may see some minor clipping at start up; running at a reduced voltage should be OK.
Thanks for responding to me so quickly, this is a fantastic thread which I arrived at after looking for information about motor controllers and the Falcon PSU.
Apologies for being a bit slow but I'm not sure I understand your response entirely, hence a couple of questions.
Referring to the R=16²/Power equation, is Power referring to the motor's requirement?
Also, if the TDA7492 amp will clip at 16VRMS when bridged, would I be better looking at some of the 100W per channel boards?
Apologies for being a bit slow but I'm not sure I understand your response entirely, hence a couple of questions.
Referring to the R=16²/Power equation, is Power referring to the motor's requirement?
Also, if the TDA7492 amp will clip at 16VRMS when bridged, would I be better looking at some of the 100W per channel boards?
Sorry for being a little late on the reply, but early bench testing of the Parts Express TPA3116D2 amp is positive.
I am running the amp off of a 16V/3.3A DC power brick from an old laptop. The input pot is a little sensitive at the lower levels, but I did get it set up for a stable 4Vrms output driving both coils of a (supposed - see below) LP-12-1900 in parallel. I don't have a working oscilloscope at the moment, so I am keeping the output voltage a little lower to avoid possible clipping.
The only really odd thing I am dealing with are the output transformers. They are both labeled as LP-12-1900 and purchased from Mouser. With each of the input coils being driven with 4 Vrms, I am seeing ~60 Vrms on each coil of the output. This is with an 11.7K resistor on the output side to act as a load. So it appears that my transformer has a winding ratio of 15:1 rather than the 18.25:1 (approx) that I should be seeing with the LP-12. The only thing I can think is that they mislabeled some LP-16 series as LP-12. Both transformers behave the same.
Anyone else experienced this?
-bill
The winding ratio sounds correct for the LP12. In parallel, the secondary is rated for 6.3VAC @3.8A. In order to maintain the rated voltage with that load, the unloaded voltage will be ~7.67VAC which will be your actual turns ratio: 115/7.67=15 it is NOT 115/6.3.
This load regulation works against you in both directions. In order to get 115VAC out UNLOADED, you will need to provide 7.67VAC into the secondary. To maintain 115VAC output under load, the input to the secondary will need to be even higher.
I used the LP12-1900 in the OP because I was using an LM3886 amp which is single ended. If you are using a bridge amp like the TPA3116, you effectively double your voltage which makes everything easier from an impedance standpoint. The output xfmr also transforms the load impedance seen by the amp as a square of the turns ratio. If the load is 1200R and you are using a 6.3V xfmr, the load seen by the amp is ~5.33R (1200/15^2). With a 12V xmfr, the load seen by the amp is 20.4R (1200/7.66^2). If you are using a bridge amp, look at post #47 in this thread for a schematic to follow. It uses an Amgis 12V toroid and a 24VDC supply and it works very well.
This load regulation works against you in both directions. In order to get 115VAC out UNLOADED, you will need to provide 7.67VAC into the secondary. To maintain 115VAC output under load, the input to the secondary will need to be even higher.
I used the LP12-1900 in the OP because I was using an LM3886 amp which is single ended. If you are using a bridge amp like the TPA3116, you effectively double your voltage which makes everything easier from an impedance standpoint. The output xfmr also transforms the load impedance seen by the amp as a square of the turns ratio. If the load is 1200R and you are using a 6.3V xfmr, the load seen by the amp is ~5.33R (1200/15^2). With a 12V xmfr, the load seen by the amp is 20.4R (1200/7.66^2). If you are using a bridge amp, look at post #47 in this thread for a schematic to follow. It uses an Amgis 12V toroid and a 24VDC supply and it works very well.
Last edited:
Well, duh. In my defense it's been a while since I worked with transformers - several decades in fact. I didn't think (ie. remember) the loss under load would be quite that high. In addition I had never really taken the time to look at the specifics of Class-D amps before, so this has been a great learning experience for me.
Right now I am running the SG4 in 2-phase mode using a single amp board. Each channel drives both secondary coils of one of the transformers in parallel. I figured if I encountered any issues with load I could drop a second amp board in and have each channel drive a separate winding like your example in post #47, effectively doubling the load seen by the amp.
This is going to be used for an older Rega Planar-2 with the original 115v motor, so it only requires a few watts once it reaches speed.
Thanks for all the help, this has been a great little project.
-bill
In my defense it's been a while since I worked with transformers - several decades in fact. I didn't think (ie. remember) the loss under load would be quite that high. In addition I had never really taken the time to look at the specifics of Class-D amps before, so this has been a great learning experience for me. Right now I am running the SG4 in 2-phase mode using a single amp board. Each channel drives both secondary coils of one of the transformers in parallel. I figured if I encountered any issues with load I could drop a second amp board in and have each channel drive a separate winding like your example in post #47, effectively doubling the load seen by the amp.
This is going to be used for an older Rega Planar-2 with the original 115v motor, so it only requires a few watts once it reaches speed.
Thanks for all the help, this has been a great little project.
-bill
Dear Pyramid
I am sorry not to answer sooner, but had urgent surgery, hence the delay in answer.
I have no ripple on the DC line feeding the TDA power amplifier. Or at least it is in acceptable levels. The filtration after the 7824 regulators (x2 in paralel to achieve 2 amperes) is performed with 2200 uf, 1500uf, 1000uf, 100nf (Rubycon) capacitors and all electrolytic caps are backed up with polyester caps.
What I have found is that if i start the controller in factory reset mode and the voltage is at 128 I read exactly 234 volts stable on the output - absolutely no deviation. Once I decide to reduce the frequency from the default 50hz to 49.88, the voltage starts to deviate. The deviation is also higher if I reduce the voltage to 109 to achieve 200 volts AC on the output. In the same time I have stable voltage on 45 mode.
I use toroidal transformer rated 35va for the output.
Measuring the frequency - I read exactly the same frequency as SG is showing on the display. So the frequency is obviously fine, but the voltage deviation is extremely strange to me. As I said the DC voltage both to SG4 and Power amplifier are in steady levels.
The eddy brake has been removed before starting to use a controller. Recently build a tachometer (thank you for this project as well) and noticed that the RPMs are not as steady as I would like to be. Hence I start measuring the output from the controller.
I have no oscilloscope at home but can borrow one (only need to learn how to use it).
Here is a short video showing the deviation in voltage (49,88hz and 200 volts on the output).
An externally hosted image should be here but it was not working when we last tested it.
BR
Orlin
I was not able to open the video you posted; maybe a link to it on YouTube?
You do not need the large caps on the output of the 7824 regulator; they should be BEFORE the reg. Normally, you only want 1-10uFd on the output and you want ripple free DC going into the reg.
Does the output voltage deviate if you start at another frequency such as 60Hz? If your mains frequency is 50Hz and there is a problem with the regulator configuration, the 100Hz ripple will create a beat frequency with the output as it moves away from 50Hz.
You do not need the large caps on the output of the 7824 regulator; they should be BEFORE the reg. Normally, you only want 1-10uFd on the output and you want ripple free DC going into the reg.
Does the output voltage deviate if you start at another frequency such as 60Hz? If your mains frequency is 50Hz and there is a problem with the regulator configuration, the 100Hz ripple will create a beat frequency with the output as it moves away from 50Hz.
If I buy a TDA7492 amp board such as this: TDA7492 2 x 50W D Class High-Power Digital Amplifier Board AMP Board+ Radiator | eBay
is it ok to connect the negative speaker terminals together to take 3 wires to my 16v Thorens motor?
is it ok to connect the negative speaker terminals together to take 3 wires to my 16v Thorens motor?
Definitely not. The 7492 amp is a bridge type amplifier so the neg side of the output is not at ground. If the Thorens motor is an AC synch type, there should be 2 separate windings and 4 leads; the two common leads of each winding may be tied together, but you will need to separate them and determine (with and Ohmmeter) which pair of wires belong to each winding.