But there is good news too, coming back to Philips VCR capstan, I mounted it and installed. Speed deviation is not measurable (again average 1 revolution speed as Supaspen sees it). When it sits at 33.33 rpm it stays there for hours. If it sometimes changes for 0.01 rpm that just means it is at edge of measurement threshold. I thing this excelent speed stability is because of many poles, but also for uneven no of stator vs rotor poles (24 windings and 16 magnets).
Noise is barely possible to hear in completely quiet room, and only when ear is close to it.
Here is what I did for dressing and mounting of the motor:
Pulley is epoxy glued to rotor (shaft is centered in pulley hole so no worries of eccentricity) and than finished with small nail file as on picture before.
Pulleys I got machined in barrel shape so small misalignment between motor and platter shafts will be compensated, plus it always centers the belt.
Since this motor has all electronics attached to mounting plate that is PCB at same time, it was necessary to prepare 3 winding contacts to be securely soldered to UVW wires, than I removed rest of pcb with angle grinder to get motor round shape (as needed by my TT design)
UVW wires connection:
Since I don't want to loose these contacts, I sealed 3 wires with liquid metal epoxy, not shown on the picture.
Motor shaft is opened (for contact with pitch roller) and I closed that with piece of copper
Machinist made me an soft steel pipe that has internal D about 2.5 mm larger than bearing housing d.
I filled the gap with common silicone again, and fixed motor . Common silicone I found very elastic, decently dumping and very well keeping shape over long time:
Finally, mounting plate (3mm Al) is fixed on mounting steel pipe by one M8 bolt (thread for which is made in the steel pipe).
Like this bottom bushing of the motor (which is not accessible any more) will be forever swimming in oil, top bushing is accessible (and replaceable) by simply lifting rotor so it can be oiled occasionally.
These motors were originally mounted upside down with open shaft part driving pitch roller. 3 screws (holes for them visible under stator on above pictures) would fix it in VCR chassis. There was no real need for trust bearing as magnet would keep the weight of the rotor, so attention is needed here, to make sure there is trust washer, also not to loose those small pieces when playing.
I have small trust washer between rotor and upper bushing, further adjusted distance with mica insulator for TO3 transistor, hole just fitted shaft diameter.
For me , I think this is coming to the end of story, simple cheap VCR capstan motor is by far better than anything else I experiments with. Next I will finish TT (in May it will be 1 year that i started play, wanna finish before) and later play with fine tuning once accurate W&F measurement method is available.
Noise is barely possible to hear in completely quiet room, and only when ear is close to it.
Here is what I did for dressing and mounting of the motor:
Pulley is epoxy glued to rotor (shaft is centered in pulley hole so no worries of eccentricity) and than finished with small nail file as on picture before.
Pulleys I got machined in barrel shape so small misalignment between motor and platter shafts will be compensated, plus it always centers the belt.
Since this motor has all electronics attached to mounting plate that is PCB at same time, it was necessary to prepare 3 winding contacts to be securely soldered to UVW wires, than I removed rest of pcb with angle grinder to get motor round shape (as needed by my TT design)
UVW wires connection:
Since I don't want to loose these contacts, I sealed 3 wires with liquid metal epoxy, not shown on the picture.
Motor shaft is opened (for contact with pitch roller) and I closed that with piece of copper
Machinist made me an soft steel pipe that has internal D about 2.5 mm larger than bearing housing d.
I filled the gap with common silicone again, and fixed motor . Common silicone I found very elastic, decently dumping and very well keeping shape over long time:
Finally, mounting plate (3mm Al) is fixed on mounting steel pipe by one M8 bolt (thread for which is made in the steel pipe).
Like this bottom bushing of the motor (which is not accessible any more) will be forever swimming in oil, top bushing is accessible (and replaceable) by simply lifting rotor so it can be oiled occasionally.
These motors were originally mounted upside down with open shaft part driving pitch roller. 3 screws (holes for them visible under stator on above pictures) would fix it in VCR chassis. There was no real need for trust bearing as magnet would keep the weight of the rotor, so attention is needed here, to make sure there is trust washer, also not to loose those small pieces when playing.
I have small trust washer between rotor and upper bushing, further adjusted distance with mica insulator for TO3 transistor, hole just fitted shaft diameter.
For me , I think this is coming to the end of story, simple cheap VCR capstan motor is by far better than anything else I experiments with. Next I will finish TT (in May it will be 1 year that i started play, wanna finish before) and later play with fine tuning once accurate W&F measurement method is available.
Shame about the failed efforts, but you learned from them, which is what is important. With hindsight, it should be obvious that ball races are a bad idea; after all autochangers having ball races for their platters rumbled horribly. But it's easy to be seduced by the smooth feel of the ball races on a head drum motor when we should know that a sleeve bearing is better for our application. By the way, the technical name for the shape of your pulley is "crown" pulley, and as you point out, it has the valuable property of self-centring a flat belt. Well done!
Thanks ec8010,
And well, at least I (and everyone reading) know better now. Besides new generation big Papst was worst disappointment.
Only way not to make any mistake is not to do anything 🤣
Still I'm happy that cheapest and most avaliable of all tested motors is proving itself as excellent choice. Hope this learning will help other diyers with same ambition..
And well, at least I (and everyone reading) know better now. Besides new generation big Papst was worst disappointment.
Only way not to make any mistake is not to do anything 🤣
Still I'm happy that cheapest and most avaliable of all tested motors is proving itself as excellent choice. Hope this learning will help other diyers with same ambition..
You're welcome. Absolutely; you have to try things to find out, and the person who never made a mistake never did anything. I'm delighted to learn that the cheap and cheerful capstan motor is the best - not least because I have five or six VHS capstan motors almost identical to yours. I would have been tempted by the Papst. But the VHS coil connections are not all as expected. Only one has the expected wiring that can be driven easily from sine waves. That might be because I was clumsy in desoldering from the PCBs and broke a wire. Further investigation needed.
So far, I have a machined (plywood) 372mm diameter platter, 45mm thick. (Derived from Herr Altmann's ideas.) Also a Harley exhaust valve (also Herr Altmann) in bearing using lignum vitae sleeves that will be the platter bearing. That took some improvements to lathe, very careful machining, and two attempts before I was happy. I may replace the motor's single sleeve with lignum vitae sleeves top and bottom. By the way, lignum vitae source is old lawn bowls, not newly savaging a protected species. Lovely stuff to machine, but prone to unexpected voids.
So far, I have a machined (plywood) 372mm diameter platter, 45mm thick. (Derived from Herr Altmann's ideas.) Also a Harley exhaust valve (also Herr Altmann) in bearing using lignum vitae sleeves that will be the platter bearing. That took some improvements to lathe, very careful machining, and two attempts before I was happy. I may replace the motor's single sleeve with lignum vitae sleeves top and bottom. By the way, lignum vitae source is old lawn bowls, not newly savaging a protected species. Lovely stuff to machine, but prone to unexpected voids.
For motors, did you measure back EMF when turned by hand, it's so easy and revealing.
I have seen Altmann tt long time ago, fantastic bearing choice 😊.
But I don't know how will you solve shaft to platter centering challenge. He just manually centered shaft on platter, IMHO that's way too loose.
I fixed shaft into platter and then machined and polished platter in one go, holding by shaft in a really good lathe.
If you can fix the shaft strong enough so bond survives machining forces, I would warmly recommend that route. If Harley shaft is long enough, might be you can fix it trough hole in platter and make contact surface for glue much bigger, than it could be machined?
Same for ensuring sleeve is 100% vertical, he just drilled and glued. This is less of a problem, in post #3 or 4 I explained how to fix it into plinth without problems with feeler gauges, but platter - shaft must be perfect before that.
I have seen Altmann tt long time ago, fantastic bearing choice 😊.
But I don't know how will you solve shaft to platter centering challenge. He just manually centered shaft on platter, IMHO that's way too loose.
I fixed shaft into platter and then machined and polished platter in one go, holding by shaft in a really good lathe.
If you can fix the shaft strong enough so bond survives machining forces, I would warmly recommend that route. If Harley shaft is long enough, might be you can fix it trough hole in platter and make contact surface for glue much bigger, than it could be machined?
Same for ensuring sleeve is 100% vertical, he just drilled and glued. This is less of a problem, in post #3 or 4 I explained how to fix it into plinth without problems with feeler gauges, but platter - shaft must be perfect before that.
I have not tried measuring back-EMF. I'll put it on my list.
I'm afraid there are rather a lot of problems that Altmann failed to mention. One thing I discovered is that Harley valves still have grinding compound/debris on their shafts. I discovered this when I came to polish the shaft and the cloth blackened. There's no way Brasso brass polish can abrade hardened steel like that, and after I had shifted that muck, using a fresh cloth and fresh Brasso did not produce any blackening. I can't help feel that assembling a Harley engine without polishing the valves first to get that muck off wouldn't work well. Washing with isopropyl alcohol did not shift the muck - it needed Brasso.
I bought the Harley valve and valve sleeve from a Harley shop in Chicago whilst visiting some years ago. But the valve will not go into the sleeve, and even if it did, when the sleeve is interference fitted in the engine's head, it would tighten up. I imagine the sleeve needs to be fitted in the head then carefully hand-reamed. But Altmann didn't mention that. I fitted the lignum vitae in my bearing housing, then machined the two sleeves (one at each end) for a fit with no detectable play. That was not easy. The bearing will probably need to be run in for lowest friction, but this thread suggests that a little friction won't hurt.
I don't see centring the platter on its bearing as being a problem. But I agree Altmann skimmed over that one as well. The outside of my platter was turned on a lathe and the grain raised with alcohol before lightly rubbing with 400 grade abrasive whilst still on the lathe. I am confident I can achieve a smooth surface when varnished (I've done this before on smaller stuff). Given a smooth surface, I can arrange for a dial gauge to measure where the platter is as it is rotated. It is not a problem to get it within 2um of correct. I have started making a boss to grip the Harley valve that can itself be adjusted for concentricity when spun in the bearing housing, then the platter seated and adjusted on the boss for concentricity.
Birch plywood is not a very constant thickness either. So the platter was faced as well. But how stable it will be in tha face of changing humidity and temperature remains to be seen. I also have a 3kg circular granite chopping board that is exactly the right diameter for a platter. Granite chopping boards are a cheap source of low accuracy surface plates and are ideal for planishing parts; I keep one on the lower table of my pillar drill with abrasive paper handy nearby.
I'm afraid there are rather a lot of problems that Altmann failed to mention. One thing I discovered is that Harley valves still have grinding compound/debris on their shafts. I discovered this when I came to polish the shaft and the cloth blackened. There's no way Brasso brass polish can abrade hardened steel like that, and after I had shifted that muck, using a fresh cloth and fresh Brasso did not produce any blackening. I can't help feel that assembling a Harley engine without polishing the valves first to get that muck off wouldn't work well. Washing with isopropyl alcohol did not shift the muck - it needed Brasso.
I bought the Harley valve and valve sleeve from a Harley shop in Chicago whilst visiting some years ago. But the valve will not go into the sleeve, and even if it did, when the sleeve is interference fitted in the engine's head, it would tighten up. I imagine the sleeve needs to be fitted in the head then carefully hand-reamed. But Altmann didn't mention that. I fitted the lignum vitae in my bearing housing, then machined the two sleeves (one at each end) for a fit with no detectable play. That was not easy. The bearing will probably need to be run in for lowest friction, but this thread suggests that a little friction won't hurt.
I don't see centring the platter on its bearing as being a problem. But I agree Altmann skimmed over that one as well. The outside of my platter was turned on a lathe and the grain raised with alcohol before lightly rubbing with 400 grade abrasive whilst still on the lathe. I am confident I can achieve a smooth surface when varnished (I've done this before on smaller stuff). Given a smooth surface, I can arrange for a dial gauge to measure where the platter is as it is rotated. It is not a problem to get it within 2um of correct. I have started making a boss to grip the Harley valve that can itself be adjusted for concentricity when spun in the bearing housing, then the platter seated and adjusted on the boss for concentricity.
Birch plywood is not a very constant thickness either. So the platter was faced as well. But how stable it will be in tha face of changing humidity and temperature remains to be seen. I also have a 3kg circular granite chopping board that is exactly the right diameter for a platter. Granite chopping boards are a cheap source of low accuracy surface plates and are ideal for planishing parts; I keep one on the lower table of my pillar drill with abrasive paper handy nearby.
@EC8010 , In post #130 I showed back EMF measuring, so simple and showing a lot, you have to try this, just call if help is needed.
Otherwise I can see we have many similar emotions about same things 🤣
As youngster I was in motorbikes, oldtimers to restore and 1cylinder 500cc hard enduro to drive. H-D had reputation of close to garbage among enduro-cross fans, good only as vintage restoration but not to drive. Honda guy needs sets of wrenches, alan keys, screwdrivers and few torque wrenches, H-D guy needs all that double in metric and imperial + honers, reamers, different grinding paste, valve seat lathe, micrometers and gauges.... Actually you sometime need to buy 5 bushings to find one that fits at the shaft. That was long ago and I thought in the meantime they went away from quality achieved by Norton or NSU in 1950's, but seems not 🙂
Might be you buy valve from Man or Scania or some other truck?
Same with Altmann's project, plywood, stepper motor and thread belt are hardly choices I would make..... But than again, if it makes him happy, why not.
Congrats on centering shaft this way! I would probably pull my hear out before I would finish.... Machining all together would be easier, but if it is done....very good!
Granite has far better potential than plywood IMO, is it ideal material, don't know. It does resonate more than limestone, but than does it matter? It is important not to induce resonances at first place (e.g. good bearing and belt drive)....
For bushing (sleeve) yes, little drag will not hurt (I think it helps if it is constant). If you already fixed them with no detectable play but turns easy, I would leave it like that and play later with oil if needed.
Otherwise I can see we have many similar emotions about same things 🤣
As youngster I was in motorbikes, oldtimers to restore and 1cylinder 500cc hard enduro to drive. H-D had reputation of close to garbage among enduro-cross fans, good only as vintage restoration but not to drive. Honda guy needs sets of wrenches, alan keys, screwdrivers and few torque wrenches, H-D guy needs all that double in metric and imperial + honers, reamers, different grinding paste, valve seat lathe, micrometers and gauges.... Actually you sometime need to buy 5 bushings to find one that fits at the shaft. That was long ago and I thought in the meantime they went away from quality achieved by Norton or NSU in 1950's, but seems not 🙂
Might be you buy valve from Man or Scania or some other truck?
Same with Altmann's project, plywood, stepper motor and thread belt are hardly choices I would make..... But than again, if it makes him happy, why not.
Congrats on centering shaft this way! I would probably pull my hear out before I would finish.... Machining all together would be easier, but if it is done....very good!
Granite has far better potential than plywood IMO, is it ideal material, don't know. It does resonate more than limestone, but than does it matter? It is important not to induce resonances at first place (e.g. good bearing and belt drive)....
For bushing (sleeve) yes, little drag will not hurt (I think it helps if it is constant). If you already fixed them with no detectable play but turns easy, I would leave it like that and play later with oil if needed.
I will go back to #130 and take a look. Well, that's easy enough to do. It has just occurred to me that an FFT of the back-EMF waveform will almost certainly show when the rotor is correctly centred with respect to the stator because it will have a less distorted waveform.
I had not considered a valve from a big truck engine; there's a thought.
It really isn't difficult to centre stuff accurately! Honest. The key thing is to be able to twirl whatever it is whilst watching the dial gauge. You twirl for maximum deflection, note the reading, then twirl for minimum and note its reading. With the whatever at maximum or minimum, you then simply adjust (tap with hammer, or operate four-jaw chuck keys) to achieve a reading that is the average of the two readings. It's really that simple and surprisingly quick. Obviously, the dial gauge needs to be held solidly in place.
Somehow, I like the idea of the plywood platter and the time has been invested in machining it. The granite platter would be harder to deal with and require drilling a hole through the middle. I have some diamond core drills in the event that needs to be done but it would be a very slow job, probably needing some form of improvised autofeed on the pillar drill.
I had not considered a valve from a big truck engine; there's a thought.
It really isn't difficult to centre stuff accurately! Honest. The key thing is to be able to twirl whatever it is whilst watching the dial gauge. You twirl for maximum deflection, note the reading, then twirl for minimum and note its reading. With the whatever at maximum or minimum, you then simply adjust (tap with hammer, or operate four-jaw chuck keys) to achieve a reading that is the average of the two readings. It's really that simple and surprisingly quick. Obviously, the dial gauge needs to be held solidly in place.
Somehow, I like the idea of the plywood platter and the time has been invested in machining it. The granite platter would be harder to deal with and require drilling a hole through the middle. I have some diamond core drills in the event that needs to be done but it would be a very slow job, probably needing some form of improvised autofeed on the pillar drill.
Last edited:
I don't really know about plywood for this, might be it is good, I just don't know and I wouldn't choose it for TT,. I do use it for many other things, from kitchen building and other furniture to speakers, but as you mentioned, it is too unstable and difficult to get in good dimensions....
Granite is very hard, depending on type, limestone is much easier to machine.
My platter was actually square plate when I have first seen it. About 12mm hole was drilled in the middle (easily) . Mason fixed bolt in his messy lathe and fixed square plate on it with nut. The rest was epical mess of turning square into the round, but he did not care, he is full of stone dust anyway. It could be eased by using angle-grinder (and tile cutter) to shape the plate a bit before and reduce debris.
On that rough platter I installed precise steel shaft. This assembly was taken to metal machinist who finished the platter on nice machine, still dusty but much less.
I think you are thinking of making it out of platter shape that has no hole, i started with hole and machined platter around it......
To drill and machine stones is not that difficult, but depends on stone hardness. for hole I would use normal wall bit and a lot of water pouring on drill. Stone must be supported underneath with something soft and straight, don't use hammer drill.
But at the end I would machine it after shaft is installed, as I described. There will be little material to take off, easy for lathe if it can take size of d300. You can protect it from dust with some rags and plastic...
In local shops you can probably find the same limestone that I used, its sold worldwide. If you wish and want to wait forever and pay for 10 -15 kg P&P, one day Ill ship you a block 😎
Granite is very hard, depending on type, limestone is much easier to machine.
My platter was actually square plate when I have first seen it. About 12mm hole was drilled in the middle (easily) . Mason fixed bolt in his messy lathe and fixed square plate on it with nut. The rest was epical mess of turning square into the round, but he did not care, he is full of stone dust anyway. It could be eased by using angle-grinder (and tile cutter) to shape the plate a bit before and reduce debris.
On that rough platter I installed precise steel shaft. This assembly was taken to metal machinist who finished the platter on nice machine, still dusty but much less.
I think you are thinking of making it out of platter shape that has no hole, i started with hole and machined platter around it......
To drill and machine stones is not that difficult, but depends on stone hardness. for hole I would use normal wall bit and a lot of water pouring on drill. Stone must be supported underneath with something soft and straight, don't use hammer drill.
But at the end I would machine it after shaft is installed, as I described. There will be little material to take off, easy for lathe if it can take size of d300. You can protect it from dust with some rags and plastic...
In local shops you can probably find the same limestone that I used, its sold worldwide. If you wish and want to wait forever and pay for 10 -15 kg P&P, one day Ill ship you a block 😎
With your mechanical and electrical (winding coils) skills and available tools (mainly dial head), and if I would have all the time in the world, I would think of making own DD TT motor.
Say 32 small air coils (slot less, no metal core) attached to plinth and 24 small magnets installed in platter by help of dial head, all on diameter of some 250mm... What can be wrong with that?
There is a way to reduce the rumble.
Get a silicone type thin TO-3 transistor insulator, cut it to fit then place it between the platter bearing washer and platter.
I did just that for an RCA 45 player and it reduced the rumble by a decent amount.
That's keen. But you would ideally need a dividing head to do it. You could make the stator for the coils by 3D printing, so that wouldn't be too bad. A twirly jig thing could guide the wire onto the coils, but as you've seen from commercial motors, getting an even wind is tricky. I suspect the magnets need to be spaced apart by their diameter and they would need to be accurately spaced, hence the dividing head. I suppose one could again use a 3D printer to make the rotor having holes for the magnets to go into. My feeling is that capstan motors will have had serious design effort put into them and are probably as good as it gets. After all, a lot of them were made and it was worth refining their design.
I'm a big fairy when it comes to the kind of swarf that I will allow on my machine tools. I don't even like cast iron because its swarf is abrasive. When you start rasing clouds of dust everywhere it settles on the (ubiquitous) oil and forms grinding paste, ruining tools (especially if it gets into lead screws). So thanks for the limestone offer, but no thanks. As it was, I had to borrow a friend's gap bed lathe to do the plywood platter. I really just want to make a turntable from scratch to prove to myself that I can. As a teenager, I made electrostatic headphones, and I now know how to do that better, so a system that was home-made from cartridge onwards (don't fancy attempting that) would be nice.
If only I had known about your silicone insulator trick when I was a teenager with terrible turntable (and reflex loudspeaker).
Last edited:
I thought you have one...
Oh yeah, it does that! there is way to reduce it a lot by putting cutting knife in water flow, but of course one needs to channel water out in neat way too...
This is completely right feeling IMO 🙂. But they don't exist in full platter size except for DD machines, and than tuned for low mass platter... Just wondering if really large but low flux motor could be designed.... Yet that's for another life, too much of an project.
Motor amplifier:
In the meantime I run this capstan motor 24/7 few days, with test rig amp, still all ok and I went ahead with final motor drive solution.
The motor likes (all per phase):
for 33rpm 2.8V RMS @ 0.3A, Supaspin at 70% amplitude
for 45rpm 3.3V RMS @ 0.3A (I listened Pyramids advice and adjusted for constant current), Supaspin at 80% amplitude
Startup at 100% amplitude which is 4 V RMS
4V RMS maximum equals 11.28V p-p, so I have chosen Vps 12VDC. This is very convenient as 1.5A external HDD PS is perfect.
Like this safety earth is not connected via PS avoiding one big potential ground loop.
Here it is important for me to choose lowest VPS that does the job to avoid extra heat inside TT housing. As mentioned before, if amp goes bit into clipping at start up, doesn't matter at all.
For such task, old fashion AB class 4 channel car audio amp is perfect.
Before mentioned L272 power opamp with 0.7A max output current would probably do to, but to much at the edge for power dissipation.
World of caution: some of those 4 channel car chip amps have 2 amps in inverting mode (to ease bridging) , at first I got STA540, that one is like that. Manageable to invert one channel but no need as other chips exist.
Final selection is TDA8562Q, about 6 Euro device. It has 4 equal amps with gain fixed at 20db (10x).
I studied Supaspin output a bit, modified 2 things and came up with this that works very good last 2 days, only 4 components; chip, 2 PS capacitors and Vref capacitor:
I deliberately copied Supaspin and chipamp diagrams to allow comparison, my amendments are in red pen:
1. On Supaspin I changed R3, R4 and R14 to 9k1 to get 400mV at outputs (with gain of x10 it gives me 4VRMS @ 100% amplitude for motor)
2. Removed R9, R10 and R17, they were referencing output DC to GND, I left outputs floating for chipamp as it has its own reference V, Which appears to be 2.1V DC. Since I measured Supaspin DC before capacitor at 2.5VDC I did not reverse capacitors (C4, C5 and C10)
3. On chipamp I did nothing. Outputs are at 5.9VDC with 12VDC PS. DC doesn't matter for motor as its not looking at ground. Max DC offset between 2 channels measured 30mV, motor will not care for that.
4. In addition I added 1R resistor inline with PS for all (this 12V powers amp and supaspin) . I use it to monitor total current without disconnecting wires and it drops VPS to bit above 11V, which is cool. I will keep it in final installation.
In the meantime I run this capstan motor 24/7 few days, with test rig amp, still all ok and I went ahead with final motor drive solution.
The motor likes (all per phase):
for 33rpm 2.8V RMS @ 0.3A, Supaspin at 70% amplitude
for 45rpm 3.3V RMS @ 0.3A (I listened Pyramids advice and adjusted for constant current), Supaspin at 80% amplitude
Startup at 100% amplitude which is 4 V RMS
4V RMS maximum equals 11.28V p-p, so I have chosen Vps 12VDC. This is very convenient as 1.5A external HDD PS is perfect.
Like this safety earth is not connected via PS avoiding one big potential ground loop.
Here it is important for me to choose lowest VPS that does the job to avoid extra heat inside TT housing. As mentioned before, if amp goes bit into clipping at start up, doesn't matter at all.
For such task, old fashion AB class 4 channel car audio amp is perfect.
Before mentioned L272 power opamp with 0.7A max output current would probably do to, but to much at the edge for power dissipation.
World of caution: some of those 4 channel car chip amps have 2 amps in inverting mode (to ease bridging) , at first I got STA540, that one is like that. Manageable to invert one channel but no need as other chips exist.
Final selection is TDA8562Q, about 6 Euro device. It has 4 equal amps with gain fixed at 20db (10x).
I studied Supaspin output a bit, modified 2 things and came up with this that works very good last 2 days, only 4 components; chip, 2 PS capacitors and Vref capacitor:
I deliberately copied Supaspin and chipamp diagrams to allow comparison, my amendments are in red pen:
1. On Supaspin I changed R3, R4 and R14 to 9k1 to get 400mV at outputs (with gain of x10 it gives me 4VRMS @ 100% amplitude for motor)
2. Removed R9, R10 and R17, they were referencing output DC to GND, I left outputs floating for chipamp as it has its own reference V, Which appears to be 2.1V DC. Since I measured Supaspin DC before capacitor at 2.5VDC I did not reverse capacitors (C4, C5 and C10)
3. On chipamp I did nothing. Outputs are at 5.9VDC with 12VDC PS. DC doesn't matter for motor as its not looking at ground. Max DC offset between 2 channels measured 30mV, motor will not care for that.
4. In addition I added 1R resistor inline with PS for all (this 12V powers amp and supaspin) . I use it to monitor total current without disconnecting wires and it drops VPS to bit above 11V, which is cool. I will keep it in final installation.
Connectivity between SS and amp looks good, the 60k (I think, bit fuzzy) input impedance of amp is right in the ballpark for settling.
Don't forget you can use the recently added 'frequency-dependant-amplitude' feature for a baseline compensation of curent vs frequency - with your 2.8v@33 / 3.3v@45 figures, you're dropping ~15% amplitude from 45->33, so at notional zero frequency the amplitude would be 100% - 15%.45/(45-33) = 43%, so choosing an FDA value of 40% will be pretty close. This would give you a 100% amplitude of 4v@45 / 3.36v@33, and the reduced amplitude value would then be relative to these values. Choosing 80% for each would give 3.2v@45 / 2.7v@33.
Don't forget you can use the recently added 'frequency-dependant-amplitude' feature for a baseline compensation of curent vs frequency - with your 2.8v@33 / 3.3v@45 figures, you're dropping ~15% amplitude from 45->33, so at notional zero frequency the amplitude would be 100% - 15%.45/(45-33) = 43%, so choosing an FDA value of 40% will be pretty close. This would give you a 100% amplitude of 4v@45 / 3.36v@33, and the reduced amplitude value would then be relative to these values. Choosing 80% for each would give 3.2v@45 / 2.7v@33.
PS, tried to make bigger post this evening several times, always Oppssss message, guess busy Saturday evening among DIY-ers, better get out 🤣
So, this is attempt with message without pictures to see if this works
So, this is attempt with message without pictures to see if this works
Ok, so pictures are issue..... Now its Sunday morning and pictures again return Oops message..... Ill try later