SARIS H4 Smart Trainer: Details // Tested // Hands-On Review

[@mikeprytherch8875]

With the Zwift trainer cost, all top end trainers like this need to be perfect, but who really cares about 25% gradient or 2300 watts, they are chasing the wrong thing in my humble opinion

[@gplama]

Exactly my thoughts too. 👊🏼

[@usman5140]

Most cyclist can’t even hit 1000

[@izatt82]

Not a single one of these trainers make me want to leave my kinetic rock and roll. So I have to manage my own power output with gear shifts don't care. Maybe once I hit 300 watt ftp it might matter? Lol at 200 now so a lot of life left.

[@yorkshirebrad]

I've used a H3 with zero issues since the day I plugged it in -- I was worried about a future bike / 12sp upgrade so nice to see your first problem was fitting the cassette, haha. i hope Saris fix it for future because i have trust in the brand after the last 2 years of the H3

[@stephenbetley9596]

Good lad. I want Saris to succeed. I started with them on my turbo journey many years ago with several wheel-on variants. Their customer service was a match and more to Wahoo. That the unit dimensions haven't changed to fix the axle issue of the H3 is a big red flag unfortunately.

[@gplama]

I didn't dig too deep into what changed for the '12 speed compatibility' but they've shifted the cassette out ~3mm. The body of the trainer remains identical to the H3 as far as I can tell.

[@stephenbetley9596]

@@gplama Changed the axle dimensions no doubt, but in reality the whole shape of the unit needed to be addressed. They could also have been somewhat progressive in making the unit forward compatible with climb/rizer etc. Sinking feeling when I saw pictures of the unit on release.

[@JohnathanFreter]

Was hoping the H4 would have a hardwired or WIFI option like Wahoo but this update makes me more than happy to stick with my CycleOps Hammer on the publicly available Beta Firmware to aid in eliminating the over read on the sprint.

[@WowRixter]

With the conglomerate that bought Saris, that has no bike equipment in their portfolio, the future doesn’t look good for Saris, though better than it did 6 months ago

[@abmetta66]

Just a note, I fit a 12sp Shimano ultegra cassette on a hunt 54 aerodynamicist and it works fine. Perhaps it’s a newer hub? Keep up the awesome work.

[@gplama]

This was a hub on an older set of 4050.

[@charlespackwood9219]

I’m surprised. I thought the h4 would be better. But it’s worse for the moment but as for my h3 it matches my power2max ng eco almost perfectly.

[@darpmalone2064]

Great video as always. I noticed that you tested the H4 against the Magene P505. I assume that to mean that it gives reliable figures. Do you have plans to publish a full review of the P505? Thanks

[@FreddyF1977]

Yes interested to know on what shimano crank a magene p505 fits on!

[@adam1885282]

I love my H3, though I got it for a song with the recent sales blitz prior to them being acquired

[@mbrzozka5053]

My saris h3 also lowers the power by 6-8% compared to the favero assioma dual.

[@matthewblue7839]

Could you possibly do a simple recommendation by price: say under $500, 5-1,000, 1-1500 and plus 2000? You provide a lot of insightful detail, but I get a little lost in what’s really impotent ? Thank you !

[@maxab7e63s5]

My H3 is three years old and still going strong. To bad they did not improve with the H4. Adding more power and gradient is not much of an improvement.

[@TakumaHabu]

I bought the H3 in the summer during the fire sale. I saw the same issue with power reading from the trainer being too low (relative to my Vector pedals) to the extent that I was unable to complete ERG workouts (because they were harder than they should've been). I contacted them and they just told me that you can't use another power meter at the same time as the trainer... I see no logic to their response... I wonder if this is an issue in trainers manufactured more recently...

[@icurt06]

I had the same issue on my H3. I turned off my power meter and connected my garmin to the H3. Game changer for my training.

[@christopherbeattie263]

Wow okay that's interesting. I remember using a H3 a coupe of years back before I ended up pulling the trigger on a KickR. It was a mates and I was always suspicious of his online performance given his IRL performance. I tried a few sprints and ended up deleting the rides because there was no way I was going to ever match the figures from that trainer. 5 second peaks of over 1400 whereas on a fresh day and going all out I'm lucky to do 1250.

[@cameronroberts5882]

I’ve got a H3 and had the same issue fitting the 12sp cassette on. Eventually got it to fit. Rear mech cage just touches the frame of the trainer but works fine. My Sram red 11sp rear mech also touches the frame of the trainer.

[@gplama]

I could have played 'arts and crafts' with the Dremel to sort out the freehub... but... I really didn't want to have to resort to that.

[@cameronroberts5882]

@@gplama I didn’t need to go that far. Just a good whack with the fist and it went on. But 100% agree it doesn’t fit like it’s supposed too.

[@neilallonby9310]

@@cameronroberts5882 problem could be getting that cassette off again….?😮

[@craigjones4372]

I would never buy a Saris H4 after my experience with the Saris H3. Saris customer support is none existent. I bought a Tacx Neo 2T to replace my Saris H3. The Tacx Neo 2T has got to be the best trainer on the market. Well worth a few extra dollars.

[@jmp485]

Thanks!

[@saracen888]

Hi Shane. In the data analysis, the pedal image is of the XC200 whereas the pedal system shown in the top right corner is the RS200. Which Garmin pedal did you use and have Garmin ironed out the issues you initially discovered (temperature and zero offset issues)? Cheers

[@gplama]

I’d swapped the pedal bodies over to the XC, the ANT ID remains the same so the analyser still shows the RS (that’d I’d assigned to them when first using them). The new firmware on them addresses issues with auto calibration during wild temperature changes. I can still make them drift with hard sprints.

[@Rob-fx5yn]

Looking forward to that magene review

[@AmericanGunWorx]

Hey Shane, not gonna lie. I used to watch you a lot, really like your video's. Several years ago I dove into cycling pretty hard and very happily. Life commitments have given me the excuses to pull away. Never the less I pulled my old but practically new H2 out of the closet and was remembering why I did not use the trainer. From day one the H2 made an annoying grinding noise which would bother me the entire time I would ride. Was this a common problem with the H2? Is the 3 any better. Back then they were under the CycleOps name and the company did not do much to help resolve the issue. I had disassembled many times to inspect. Do you have a good suggestion for a reliable trainer. Don't care about power output just want an enjoyable ride and exercise maybe some interactive software. Did not really care for the Rouvy to much. Maybe things have gotten better. Thank you for any input, always enjoyed your content. Mark

[@gooders9093]

I have had a Neo 1 since they first came out. Always looked after but is it time for a change. If I do I ain't buying anything until it's got an A rating from you.

[@marcelgreiner2844]

I think the Spindown calibration is a Zwift problem since it stopped working with my Kickr Core as well. It doenst read power and then justs stops the calibration.

[@giodc8599]

I recently took a gamble and fitted a Saris XDR driver with a 10-36 cassette from Rival eTap. It works! And my derailleur cage is like 1-2 mm from the body of the trainer so it's ok. Not sure if the cassette went all the way, it was a bit funny how i had to fit it on and it felt like it was meant to go further anyway. Any idea?

[@kidsafe]

SRAM 12-speed road has narrower cog pitch than every other 12-speed implementation. The flattop chain is 5.0mm wide while Shimano chains are 5.25mm. If we assume that width difference is also the difference in cog pitch, then a 12 speed SRAM road cassette could be 2.75mm narrower than a Shimano one.

[@giodc8599]

@@kidsafe My setup is 10-36t on both the trainer and my actual bike. The combo works though i get some rumbling on the 11t and 10t cog. Not sure if the cassettes are ever so slightly differently aligned between bike and trainer or just the way it is. And i got a Saris H3.

[@mrwezbo]

Interestingly it doesn't work for me on a H2, 10-33t force. RD hits the trainer body and I cannot shift into the 30t/33t

[@kidsafe]

@@mrwezbo Lama mentioned the H4 seems to have shifted the hub axle a couple millimeters to the drive side, sacrificing some clearance on the NDS. The NDS part of the enclosure is now stamped for more disc caliper clearance as a result.

[@lifeincycling]

Oooh cog science!

[@gplama]

Pure hackery on my behalf... but I was able to eliminate 'user error' as the problem. That's always my first assumption. :)

[@mariskoralkevics9675]

My Kickr Core also shows about 5-10% more than Garmin Vector 3 dual. I do factory spindown and it doesnt help. Sometimes it shows nearly similar numbers, but mostly inaccurate. I train at 10'C and think flywheel isnt overheating.

[@davekashuba4730]

If you have a friend nearby with a power meter on his bike, I'd put his bike on there and compare the results. That what is frustrating about power meters - when 2 are different, it's hard to know which one is correct. My Core always reads low initially, then after 10 minutes is spot-on with my Quarq Elsa RS and 1-2% higher than my NGEco. The warmed-up Core is also spot-on with my now defunct Elite Direto. If you prove that the Core is reading high, then Wahoo's customer service is pretty good - I'd reach out to them with your findings.

[@briandarnell1809]

I'd bet the pedals were off. I've tested my Kickr with a Quarq D-zero and Garmin CX200 pedals. The pedals are outliers. Kicks and Quarq almost identical. I've found the pedals way too temperature sensitive.

[@madsbay360]

I have a Saris H3 (i think) and my 9 speed is very hard to put on. So might just be how they're made?

[@Papa-Bogey]

(Me, smiling and looking at my Elite Direto X) ...

[@sam2288ful]

Is the lama back in the rattatat yet

[@modarm]

Talking about cadence, I have seen plenty of people complaining about Kickr 5 reporting low cadence situation being out upto 10% in worst cases against cadence sensors. Have you had this Shane with your Kickr units.

[@gplama]

It really depends on the rider/pedal stroke. I find the Kickr overestimates a few rpm in ERG. My recent Kickr 6 video showed this in the data review.

[@timeanspace]

Yep, still doesn’t work with big cassettes or through axle properly. I moved to a kickr from my original hammer because of that and I’ll never be going back unless they completely change the form factor

[@njm3211]

Very disappointing actually. I have had an H3 since the start of Covid. Generally pleased with the exception that the Ultegra long cage derailleur clearance at the lowest gear hits the plastic case. How stupid is that? One has to wonder who designs and tests their prototypes prior to final production.

[@larrywhite8590]

I may be in the minority but trainer absolute accuracy is not really critical to me. If you get within a few percent all good. What matters to me in order is 1. Ride feel, 2. Erg ramp performance, and 3. It has to be quietish. Why? Like a lot of folk I use an on bike power meter, I carry this over to Zwift for consistency. Riding a trainer is not “fun” per se so ride feel better be right. Noise? This obsession kind of amazes me, as long as we aren’t talking the Lemond revolution then whatever. With the fans and drivetrain no trainer cave will be quiet.

[@gplama]

The power at least needs to be within spitting distance of what they put on the box. It’ll get companies in trouble with the ACCC here (government consumer protection body) if this was ever officially challenged.

[@brent4adv]

I would say most folks use the trainer power meter when on Zwift. I don't have a power meter on my bike so I use the trainer power when I ride on Zwift. With that said, I'm not super critical of accuracy since I don't train with power outdoors. However, if you're going to market 1% accuracy then you'd better be able to back it up.

[@larrywhite8590]

Don’t get me wrong I think the specs need to match reality but the obsession about 0 db and 0.5% accuracy really kind of amazes me. Ride feel and ramp performance seems to be way down the list, but they really affect the experience.

[@gplama]

@@larrywhite8590 As above, it's illegal for companies to make false or misleading claims about their products in most countries. Ignoring or dismissing these inaccuracies is a leave pass too many people give out. It's why companies think they can keep churning out shitty products at a premium price. This industry needs to step up. And to do so, they have to be held accountable. We're 10 years plus down the path of direct drive smart trainers.

[@galenkehler]

still baffles me that we use standard cassettes on trainers at all. 1. weight is largely irrelevant, so why are there not ultra robust super hard wearing gears? 2. we have ability to vary resistance, why are still shifting at all, wearing chains and derailleurs, shifters and cables. 3. power accuracy is dependent on flywheel speed, so why are trainers allowing such a variance in flywheel speed? a single gear would be more accurate. 4. just thought of this, but a single speed would also eliminate cadence inaccuracies as you could clock the cadence to the flywheel speed.

[@gplama]

Smart bikes are already onto you! 😀 Belt driven, single gear, etc. We just need to see that trickle down (across?) to direct drive smart trainers.

[@irvhh143]

The pros want a trainer they can take to an event and use their race bike on. The question is why do we need 12 spd when they can't even make enough bikes.

[@gplama]

The pros have someone haul trainers to events for them... or they just throw a Feedback Sports Omnium in their bag.

[@kidsafe]

On my SB20, flywheel speed decreases if I change to a harder virtual gear and increases when I change to an easier one. That part makes no sense to me. Flywheel speed should not be affected by gearing changes, but gradient and power changes while running apps like Zwift. The SB20 is just a whole other can of worms I could go on about. I should not be able to affect power measurement with non-rotational forces (squeezing my knees in or pushing them out,) but that’s precisely what happens. Plus the power measurement is noticeably affected by a battery’s supplied voltage.

[@ralfmimoun2826]

Real bikes have deraillers. How do you want to eliminate the risk to destroy them when shifting on a "non-shift" trainer? And no, don't want to buy a single speed bike. With one cog, you have drastically different flywheel speeds, from 100W up to 1500 (and that's the range for hobbyists)

[@DavidMulligan]

Could drivetrain losses explain the differences?

[@gplama]

Not easily. Loss doesn’t explain the few sprint overshoots.

[@kidsafe]

😬

[@pierrex3226]

Reminds me of Ron Swanson praising capitalism as kill or be killed (the platter for 12 people scene). An inferior product / company in a competitive environment will fail. And while people get easily convinced to change bikes every two years, you only need one smart turbo trainer, a market that Tacx saturates, with a few others. Hence, why would anyone buy this?

[@bluesthemoose]

Well, another one to add to the list of products that don’t work you’ve been testing in 2022 🤣

[@gplama]

..... and I haven't even scratched the surface on here of "shit that just doesn't work" that I've come across this year. I'd love to do happy reviews... but that's entirely up to these companies to make better products. Or at the very least, products that at least work as they say they do.

[@bluesthemoose]

@@gplama you should make a “s**t that don’t work at all” video… nothing too detailed, just a funny one to get it off your chest and give us a good laugh!

[@gplama]

@@bluesthemoose That might have to go on 'Members Only'.... or OnlyFans! 🤣

[@bluesthemoose]

@@gplama I’d pay good money for that (well, as good as the Euro can be these days…)

[@jakefrommalibu]

So you’re ripping a product that isn’t being shipped yet. They’ve acknowledged the issues you describe. You speculate that they aren’t shipping because they haven’t solved the problems. If they eventually ship without fixing the problems, a video like this is useful. But if they never ship, this video is a waste of time. And if they fix the problems before shipping, it’s even worse because you’re describing issues that don’t exist. At best, you should have postponed publishing this until they ship . At worst, you’re publishing wrong information.

[@gplama]

Interesting take. This isn’t a ‘pre release’. This is a production unit. Saris had four weeks to clarify this wasn’t representative of what is shipping. They didn’t. They had every chance to ask me not to review a unit they sent for review. They didn’t. If you want happy reviews and actual technical detail brushed over, there’s a few other places I can point you.

[@All4Grogg]

If the spindown is anything like my Saris wheel on trainer, he was being VERY generous. He works for us, the consumers, not Saris. Honesty is non-negotiable. No way should any of this be a surprise to them, if it is that is a huge process problem and the new ownership had better make sure these weren't known, reported problems that were shoved out the door because some idiot at corporate decided to ignore them. Having spent the last 15 years in QC, I'm very, very jaded past the point of accepting self proclaimed blind ignorance to this kind of thing. The product failed to meet the published specifications, that right there, is a serious problem and one that could easily destroy their business. Shane is being fair and honest and I'll bet there are more than a few people who've worked on this that appreciate the candor and others who'll be spending the next weeks fighting the urge to scream "I F'ing TOLD YOU" to some sales/marketing/executive @%#$ that ignored them because they weren't saying what they wanted to hear.

[@kristianjorg418]

@@gplama Looks like a misconfiguration in the software with the power output. Should be possible to fix by Saris with a firmware upgrade. If that is the case, it is a no problem. Software issues are easily fixed. Hardware issues are worse. The hub problems can be remedied with a free replacement unit IMHO - if the customer needs it. Will be a bit costly for Saris though. Another way is to use a dremel to grind away a tiny bit on the cassette. Not ideal, but a simple workaround (and one I would use). The stick problem with the casing beeing to close is a stupid thing though.