Several things could be going on, not all of them in the trainer. If your hub bearings are getting hot due to pressure or perhaps are adjusted too tight, I could see increased resistance when you are pushing harder (pressure in the bearing systems of your bike) or dirt in the bearings causing similar resistance to rolling.

So If you have the mechanical chops to tear down your fluid trainer, I think you have the chops to give your bike an overhaul, clean, regrease, renew bearings as may be necessary, and readjust the bearings so as not to have excessive bearing pressure. I tend not to use preload in my bearings, I would suggest giving that a try. But you still should not have any play in the bearings.

It's still of value to finish overhauling your trainer, but don't neglect renewing the hub, BB and perhaps freehub bearings.

The theory here is, with all known good main bearings, there should be zero change in rolling resistance with pressure, speed, heat buildup, or running time. But I could imagine standard BBs causing such trouble if they are not up to snuff such as dirty, grease not fresh, worn, or too tight.

The rear bearings felt beautiful when I checked them but since I was in the process of going through the bike, I did clean & lube everything. I like to set my bearings so there's no play and there's no sensation that the bearings are being pressed into the race by the cone- just smooth rotation. The bearings had almost no lube, but the wheel turned beautifully.

The trainer isn't clamped so hard on the skewer than it could cause a problem and the hub isn't cranked into the tire too hard.

The OEM Suntour crankset has worn rings but that's not where the pulsation comes from- the OEM bottom bracket has sealed bearings and they're in good condition, but I did change the crankset and BB because of the wear. The chain is good, clean and lubed- I tore the freehub apart to clean & lube it- like the wheel bearings- little lube, no play, no preload before clean & lube.

I talked with my friend who gave the trainer to me but I forgot to ask if he did anything to it (like change the oil). Since the oil isn't used for lubrication, I think I'll empty the reservoir and reassemble it, so I can compare. The bearings are sealed and separate from the impeller housing, so this shouldn't cause any problems for them. If it eliminates the pulsation and added resistance, I'll see if I can find something else to fill it with.

Pulsation is more likely to come from the rear tire/wheel, though it should be easy to tell from the pulsation's frequency. In fact, frequency should be a dead give-away.

Shouldn't it do that constantly? It doesn't- only starts after about 15 minutes and only if I'm in a higher gear, riding fairly hard. I'm not standing up because I'm still rehabbing my knee, so I'm not really going all out. It always stops if I slow or stop pedaling.

Once I finish clean/lube and setup for my road bike, I'll try it with that one to see if it still happens.

Resonance
 

Two thoughts: I have a magnetic (cheap) trainer. 1) In the course of my fiddling, I left the resistance unit loose on its pivot once. It made no difference at first, but at a certain speed, there was a sharp vibration which smoothed out again at a higher speed. Tightening the pivot was the cure. 2) Your idea of running it dry as a test sounds good. What about using ATF as the fluid? I'd be surprised if there was anything magical in the base chemistry of the oil. You could try various viscosities.

I'd bet that the only person that REALLY knows how it's designed speaks Chinese. Anyone on this side of the Pacific is just a middle man.

My trainer is from Blackburn but they said they have no info, which may well be due to the fact that they just buy products that are designed by someone else- why go through the expense of designing something that can be bought off the shelf, right? I called Jensen Bikes and someone told the person who answered my call that these use 'Thermodynamically- neutral, medical grade silicone fluid'- I was told the same by Kinetic, but they couldn't say what the viscosity was, either. That almost makes it seem like something used for medical implants. I would think the reason for it being thermodynamically neutral is to prevent drastic expansion at high temps and blowing out the seal but with some nominal amount of air inside, that should provide some cushion for the pressure.

Search for message # 21892593. He used hydraulic oil, and it worked.

Their "description" of the oil sounds like sales blather. You're not going to eat it. I hope.

I was thinking about using power steering or tranny fluid- both handle far higher temperatures they're pretty stable and they contain anti-foaming agents. Pretty sure the viscosity would be OK, too.

I get the thermodynamically neutral part but not the medical grade silicone. I'm sure the six syllable word impresses some people but in my tech comp class, the professor called the use of a large number of words with three or more syllables a document's 'fog index'.

I'd like for someone to open their trainer and check the viscosity- they could describe it as 'Mrs Butterworth's', 'Log Cabin' or 'Generic. watered-down big box grocery store' syrup.

The silicone oil I put in mine is about the consistency of cooking oil. It's individual to the particular resistance unit and its gearing, but I get about 19mph at 200 watts. The original spec on the unit was that it was a 500 watt unit, though I don't see how they could put a number on it. Without the unit connected, I can hit almost 40 on my rollers and a fluid unit's resistance curve gets a lot steeper the faster you go. Maybe they figure that the tire/roller coefficient isn't high enough to push more than 500 watts.

Where did you get the oil?

I don't remember - that was 20 years ago. I googled. Or Amazon.

I asked the guy I got the trainer from and he told me he didn't change the oil, then I talked with someone at a place that sells & repairs machinery- decided to guy some 40wt and try that. If it's too thick or thin, I can adjust the viscosity. It's not a Lamborghini.

I filled it and left it on the bench to make sure it doesn't leak- I'll install it and see how it works later today.

Well, that seems to have been a cheap fix- I reinstalled the drive and rode it for about 15 minutes yesterday, which was a good thing, considering I still have breathing problems from COVID. I still feel the tire tread, and did seem to feel it more or less, depending on how hard I pedaled and which gear I was using, but it never rumbled or became more difficult to crank.

Thanks to all for the comments.

Bottom line for anyone who feels the need to change the oil- pay attention to the viscosity.