Skip to content
HearingTracker
Log in

How reproducible is REM? What can skew the results?

29 replies969 viewsLast reply

Jump
Link to this topic
Reply

This is probably more of a question for practitioners: I’m curious about your experience regarding the reproducibility of results. Here is why I’m asking:

Last year I got used Phonak Sphere I70s.

I do DYI programming, but my past experience with Jabra EP20s from Costco has taught me that the settings suggested by the fitting software differ a lot from the final result after doing REM. In other words, I need REM!

I found a local clinic that was willing to do REM for a very reasonable fee. Everything was great!

Fast forward to February this year: I was able to purchase used Phonak Sphere I90s and since my had worsened in the meantime, I went for another REM session with the Sphere I90s.

This time the outcome was much worse. I have a copy of the REM results, and even though it shows that everything was close to target, in my perception lower frequencies were under-amplified. Everything sounded tinny. I increased low frequency gain in Target and things were better.

Fast forward another few weeks: I managed to acquire a used Aurical FreeFit REM device for a reasonable price, so now I can do my own REM! Yeah!

For my first self-REM, I tried to adjust the Sphere I90s again. I ended up with way too much low frequency gain. At this point I didn’t know whether my REM device was calibrated correctly, or if it even worked properly. Or whether I inserted the probe tubes correctly.

A few days later, I tried self-REM again. This time the results were great! The I90s have never sounded better! No further adjustment needed. This also means that my REM device works, and its calibration can’t be too far off!

But it leaves me wondering what happened:

  1. The first REM resulted in not enough low frequency gain
  2. The second REM yielded too much low frequency gain
  3. The 3rd REM results are great

I am using Phonak 3M receivers with power domes. I push them in very deeply, that way I have zero issues with occlusion. Normally, the power domes seal very well. The addition of the probe tube does cause some noticeable leakage which otherwise isn’t there.
I produce very little ear wax, nothing was clogged up.
I did a probe tube calibration as well as and . The results were normal.

Any ideas what could cause this variability between the three different REM tests?

Thanks!

10 people in this conversation13 min readJump to latest

Remember it’s just another tool in the toolbox, definitely will never give you exact same everytime you perform , so many variables on any given day, send your Aurical in to any place that offers certification in calibration, just so you know 100%.

Thanks, but I don’t think calibration (or lack thereof) is the issue. The first was done at a clinic by an audiologist with presumably calibrated equipment. It didn’t sound good either.

Oh yeah I got that,me too, I’ve had many times over the years and it’s completely out of whack, but just for your own reference to have your Aurical machine calibrated (just piece of mind)
Probe placement is quite critical, can it be done single handed ; )

@wtolkien said:

used Aurical FreeFit REM device for a reasonable price

What is considered a reasonable price? I would buy one but there do seem to be many potential pitfalls.

Can the pros comment on which systems might be best for DIYers?

’s don’t need this tho, pushing this theme that it’s needed is making the water’s muddy, if you wanna throw money into it “just for the sake” sure why not, otherwise quite pointless, I had a quick look around, like 2 or 3k for good condition used, right up to 5 or 6k, eBay auctions I’ve seen for 5 or 6 hundred which are selling only as “parts” etc

I paid about $1300 USD. It’s just the ‘FreeFit’ with speaker, but without the ‘Aud’ audiometer. But that’s enough to do .

Yes, it’s overkill for most people, including myself :slight_smile:

But it enables me to buy used and set them up properly. As mentioned before, my ear canal geometry seems to be such that I don’t get good results from the fitting software’s default gain settings. REM is not optional for me!

I’m only in my early 50s, so over time it will probably pay for itself. Also, over the last two years my hearing has declined quite a bit. If this continues I will need more frequent re-adjustments.

You usually calibrate the probe tube with a swept tone when you fit a new one at the beginning of the session. The probe tube is looped back on to the local mic and placed next to the system speaker.

The low frequency issue is likely down to either mic tube placement/wax or vent/seal changes caused by placement or the impact of the tube at the canal wall. It’s a bit of a weakness if you’re using as the tube will potentially cause slit venting at the dome periphery. Solid moulds don’t really have such an issue because the canal wall or tube deflects more.

Placing a tube yourself can be fairly difficult as you can’t really see what you’re doing and the tip ideally wants be within 6mm of the drum.

@Um_bongo, thanks, this all makes sense. I did perform a probe tube calibration and the software indicated that tube was inserted to the correct depth.

Additional ‘slit venting’ due to the presence of the tube almost certainly happed.

Is it fair to say that power domes would have the worst outcomes compared to more open domes or custom molds?
Is there anything that can be done to mitigate this?

Many thanks!

@wtolkien said:

Is there anything that can be done to mitigate this?

Curious about this too. It seems repeating it several times while removing and reseating the probe and hearing aid might be one possible approach. That is time consuming. One still does not know which results are more accurate. Personal perception may be off.

In reply to @wtolkien

Potentially if you had a spare dome and a heated needle/pin, you could make one/pair with a small hole in both domes to give a slight interference fit on the probe tube and use it as your ‘calibration set’. This would also offer significantly better control of the probe tip if you were doing a self measurement.

I’m not saying someone should have patented what I’m saying there: Starkey got it half-right before. You could improve first fits of no end using it and the correct mic adaptor.

@Um_bongo , funny enough I actually thought about poking holes into the power domes and feeding the probe tube through them - that’s what you meant, right? Not sure if it’s feasible.

What was the Starkey solution you were referring to?

In reply to @tenkan

@tenkan said:

Remember it’s just another tool in the toolbox, definitely will never give you exact same everytime you perform REM, so many variables on any given day,

And there you have it. is necessary and the be all and end all, some claim. And yet the results vary. REM testing or no, after we walk out of the near perfect sound situation of the Audi office, and let’s say, a dome loosens in the canal, then the real world results will be quite different.

I’m about to read all the messages in this discussion, but probably there is some answers here, especially about vent compensation part:

[Target Software - Phonak Marketing Catalogue] (Target Software - Phonak Marketing Catalogue) in which is:

And:

PH_FAQ_TargetMatch2.0_210x297_EN_029-1322-02_V1.00.pdf2

In reply to @Um_bongo

@Um_bongo said:

The low frequency issue is likely down to either mic tube placement/wax or vent/seal changes caused by placement or the impact of the tube at the canal wall. It’s a bit of a weakness if you’re using closed domes as the tube will potentially cause slit venting at the dome periphery. Solid moulds don’t really have such an issue because the canal wall or tube deflects more.

Thanks for sharing. I was disappointed with my in September 2025 because there wasn’t much gain, especially at lower frequencies. I think this was probably because of the situation you described.

I was fitted with power domes that were placed very deep in my ear canal, and I even let the poke my eardrum just to make sure its tip was 1-3 mm from it.

@wtolkien , out of curiosity - have you tested by testing with 90 (85 dB) input? And I wonder whar REOG was with your power domes if you want to share it (I will understand if not).

In reply to @wtolkien

Starkey built in a type of verification with a mic port adaptor a couple of generations ago: though it relied on fitting around the side of the mould.

In reply to @Bimodal_user

@Bimodal_user, thanks for sharing your experience with not getting enough gain at lower frequencies! This mirrors what I got when the Audiologist at the clinic did for my I90s back in February.

Also, in January I had REM done at Costco with my EP20s and the Audiologist there remarked that she was measuring low frequency gain that was well above target. She had no explanation, but was smart enough to leave it as-is: “If you like it that way I’m not going to change it”.

This was probably another case of skewed results due to slit-venting with power domes!

I also did what your did for probe tube placement: insert them very gently until they touch the eardrum, then retract a bit. That’s probably not the recommended method, but the tube is so soft that I don’t think it could cause any damage.

Below are the results from my 2nd ‘self-REM’, which resulted in gain settings that sound very good to me.

Here are the REUG (black) and REOG (purple):

REUG and REOG

And these are the results after adjustments. Dotted lines are targets, solid lines are measured. Curve #1 is a 85dB sweep for (I think :slight_smile: )

REAR and MPO

I’ve been doing my own for several years. At first it was variable but for the last couple of years I get very consistent results Tube placement depth and they type of ear tip/mold used are the most critical pieces, provided you calibrate the speakers each session. I use MedRx system. Hearing aid software underamplifies my high frequencies by a lot, and they do the same with my father. Costco way undershot his Rx even with REM, though I think that was intentional on their part when he was a new customer.

Having used it more now do you recommend the MedRx for ? It seems a simpler system but whether that is accurate or better is unclear to me. It seems a bit cheaper when one can find them.

I don’t think any device manufacturer can afford to bring a system to market that is not accurate. In a professional setting they require regular calibration though.

In my opinion, for DYI it almost doesn’t matter what system you have. As a DYIer you don’t care about efficient workflow, Noah integration, 24h on-site repair turnaround, etc.
Those would all be considerations that might sway a clinic one way or another, but are almost irrelevant for DYI.

Related discussions