Kenton Thru-25 Lets You Connect 25 MIDI Devices To One Source


Kenton has introduced the THRU-25 – a box that lets you connect up to 25 MIDI devices to one source. 

The Kenton THRU-25 receives MIDI data from a single MIDI OUT (e.g. sequencer or master keyboard) and produces identical copies at the 25 MIDI THRU sockets. These can then be connected to the MIDI INs of sound modules or other devices which require MIDI signals sent to them.


  • 1 MIDI In & 25 MIDI Thrus
  • Works for ALL MIDI messages including Clock, SysEx, MTC etc.
  • Includes schmitt trigger logic for signal quality restoration
  • Less than one microsecond latency
  • DC power input jack 2.1mm (centre +ve)
  • Power/Data indicator light
  • Metal housing

Pricing and Availability:

The Kenton THRU-25 is available now for £128.92 or £154.70 with VAT.

49 thoughts on “Kenton Thru-25 Lets You Connect 25 MIDI Devices To One Source

      1. Came here for the ill-informed, knee-jerk, wrongheaded negative comments, was not disappointed.

        Way to go, Synthtopia readership! You are nothing if not predictable.

        1. Would love to know how asking what they were thinking by not making the rackable counts as “ill-informed”. A device like this is the perfect candidate for the 19″ form factor, and while it’s great they could squeeze it into 12″ it would be at least as desirable with an internal psu (rather than the wall-wart supplied) and a pair of rack ears.

          1. i agree , not because i made the initial comment but. there is no usb and it dosent fit in a standard rack . that is common sense stuff. thats my opinion and im free to type it. i think its pretty justified. and im not alone apparently.

            i just don’t like when companies take the time to produce something that could be done properly but then bail on it at the last minute . if your making a midi merger / thru and you don’t have a usb and have to plop this thing on your desk with a mess of wires hanging out of it the appeal is lost . good luck

            1. No USB makes sense to me–there’s a whole other rigmarole involved in doing USB MIDI that’s not involved with a mult, which will just be an optocoupler and a couple of Schmidt triggers at its core. Then you have to worry about timing, drivers (if you want said timing to be good), etc. however, a big box like this should really at least have a rack kit *available* if not stock, and an internal psu.

          2. The reason a lot of electronic gear is made with wall warts rather than internal PSU’s is due to C-Tick certification – at least it does in Australia.

            If something has an internal PSU it needs to pass C-Tick testing to make sure it doesn’t have adverse effects, e.g. generating radio interference, electromagnetic interference.

            If something has an external psu, C-Tick testing is not required.

          1. Putting this in a rack wouldn’t make any sense without a complete redesign, because half the connections are on the back and half are one the front.

      1. I forget who makes it but you can get a 6>1 and a 4>1… It is the same ones that make my 1>4 thru that it’s a permanent fixture in my setup. It makes sense of you have say a synth with separate voices so that you can send parallel channels w/o daisy chaining or if you have say an omni knob controller with key controllers per channel… A lot you can use it for

    1. Well, they already make a 4-port merge, but by definition MIDI merge is far more difficult than mult – you have to deal with buffering signals that come in simultaneously, and then buffer overflows if there are too many messages being generated. Most large mergers use a microcontroller, and then you might as well be making a MIDI interface/router like MOTU. 4x MIDI express XT will merge 25 separate ports! Or 4 Midi Solution M8’s, but probably not a good idea to chain those…

    1. You’re over-thinking this. A midi mult usually has extremely simple electronics, and no microcontroller or logic circuits. You’d need a MIDI interface or router to accomplish this. I use a MOTU micro express – being able to configure the routing (and all the channel muting options are a lifesaver!) with my laptop and then ‘set and forget’ is great!

  1. With MIDI only having 16 channels, what kind of uses does this have for having more than 16 outs beyond clock distribution and maybe stacking synths to play the same part?

    1. It lets studios with large MIDI rigs keep everything connected all the time.

      For example – I’ve got three MIDI drum machines that I set to channel 10. Something like this would give me enough MIDI outs that I could have all my drum machines connected and then I’d just turn on the one I wanted to record or just bring up that fader on the mixer.
      This should also be lower latency, more reliable and easier to troubleshoot than daisy-chaining a bunch of devices.

    2. > With MIDI only having 16 channels, what kind of uses does this have ?

      ex : sending transport control and sync
      ex: program change for fx units
      ex modulation
      ex: sysex
      ex : with 128 note range you could have 25 different sounds or samples ( not stacked) on ch 1 alone

  2. It’s great to see something like this on the market. Bit of a shame that I’ve already spent so much time and effort perfecting my MIDI setup, but otherwise I would definitely go for one of these as it would really simplify routing around the studio!

    1. Not really – mult electronics are usually very simple. I’d guess the variability on the optocoupler and schmitt triggers to be in the low ns range. There’s no microcontroller – hence, clocking issues to much things up.

  3. This is crazy!! I have just been thinking about how I can wire-up most of my MIDI units without having to rewire all the time. Midi Solutions also does a box like this, but I think 8 max.

  4. > Less than one microsecond latency
    is actually bad. according to
    the following oldschool products are way better:
    Roland A-880
    Max. Jitter between MIDI Clocks: Zero samples (0.00ms)
    Processsing Latency: Zero samples (0.00ms)
    MidiMan Thru 1×4
    Max. Jitter between MIDI Clocks: Zero samples (0.00ms)
    Processsing Latency: Zero samples (0.00ms)
    Roland UM-550
    Max. Jitter between MIDI Clocks: Zero samples (0.00ms)
    Processsing Latency: Zero samples (0.00ms)
    Korg KMP-68
    Max. Jitter between MIDI Clocks: Zero samples (0.00ms)
    Processsing Latency: Zero samples (0.00ms)
    instead of purchasing the kenton i would ebay some of the older stuff. nothing beats 0.00ms latency.

    1. Latency is not the same thing as jitter. And sub-millisecond latency, let alone the sub-microsecond responsiveness of the Kenton, is not audible in the realm of MIDI.

      Your comparisons seem pointless. To be meaningful, you’d want to match the Kenton up against a comparable solution, which would be hard to find. To hook up 25 devices to any of the boxes you mention, you’d end up with a daisy-chain solution, which would have its own latency and possibly jitter issues, and would probably be real unreliable.

    2. “Nothing beats 0.00 ms latency.”

      Actually, quoting 0.00 milliseconds allows for up to 5 microseconds (0.005 ms) latency. So your quoted figures are meaningless in this context.

      1. > your quoted figures are meaningless in this context.
        no, they are not. the kenton company has a poor record regarding latency. get this:
        Kenton Modular Solo – MIDI to CV, Gate and Clock Converter
        Source: MIDI Clock
        Clock Out – Max. Jitter between Clocks: 4 samples (0.08ms)
        Source: MIDI RX
        Gate Out – Max. Jitter between DC Gates: 2 samples (0.04ms)
        Average Latency: 55 samples (1.14ms)
        (scroll down)

        1. karsten

          You might want to stop quoting irrelevant numbers until you have a better understanding of what they mean.

          0.08ms is super-tight in the context of MIDI and even in most audio contexts. Take a look at the 808, which people revere, and it’s got more than 25 times as much jitter. Or that 909, which is even higher.

          You’re also comparing latency for pure MIDI devices to MIDI to CV convertors and pure MIDI devices, which makes no sense. CV will always have the lowest latency and jitter, but that’s meaningless when you’re hooking together a bunch of polyphonic midi synths.

          1. yadda-yadda. i won´t buy any kenton stuff, that´s for sure. and–oh, btw, don´t forget that hardware sequencers work differently re midi compared to daw´s. tracks in midi have timing priority. midi is serial. serial means the events travel down the wire one after the other. inside the yamaha motif workstations, for instance, first in timing priority is channel 10, typically used for drums. next is track 1, then 2, 3, 4, 5, 6, 7, 8, 9, then 11 til 16. the same goes for roland´s fa-06 and fa-08 where track 10 is also the drum track.

    3. ms means millisecond, 1/1000th of 1 second. Microsecond is 1 millionth of a second. It goes on to nano seconds and femtoseconds (getting 1000x smaller each time) but you’re very unlikely to see anything with that sort of precision outside of a high-end physics laboratory.

    4. yeah… There are some problems with that Innerclocks list. I’ve had some discussions with the ICS guys about this. They added some descriptions to help clear things up, but the methodology is still a little problematic. I understand why they’re reluctant to change, they’ve been doing their tests for a long time. However, jitter is an average, so ‘maximum jitter’ isn’t really something you can describe in a repeatable way. I agree those figures are comparable to each other, but they are less useful if you wanted to measure the jitter of some equipment you have at home. ‘Max jitter,’ as they’ve measured it, is overly sensitive to fluctuations in testing equipment, heat, small voltage peaks, etc. causing outliers.

      That’s my spiel about that list. It’s a useful list, just be careful about leaning on it too much.

      ANYWAY, I haven’t seen the electronics, but I very much doubt this kenton box introduces jitter at all. Mults don’t require a microcontroller, or a clock, or any of the things that normally introduce timing deviations. There will be some small delay from electrically buffering things, but this should be always the same, so no jitter.

      Assuming this will cause lots of jitter because a list says some other Kenton thing has some jitter is silly.

  5. I have a nice thru box made by Casio. It has two switchable inputs and 8 outputs. Very useful when I was using hardware especially because of the two inputs. I think I’ll put it up on ebay. 🙂

  6. I can see this being handy for some of us still holding on to our legacy midi modules. I have two DMC MX8s
    which I use under certain circumstances. I like the Yamaha MCS2 units. but the MX8s can rechannelize. I can’t think of any other midi processors/patch bays including Akai, Ensoniq could rechannelize. I still have some early Fatars for teaching which transmitted on only channel 1 and they are still in use since the keybed feel is pretty solid. With the MX8 they could be hooked to the hardware sequencers for multi channel projects.

    1. To comment on my own posting. I think a misdirection happened once computers came into use. It was sort of similar to what happened to classical music when theory overrode emotion. I do not use computers because they force on to engage the eye. Engaging the eye can actually atrophy the ear’s strength. I welcome arguments to the contrary

  7. I’m glad Kenton keeps doing what they’re doing. Does seem a lost opportunity to not have another input (or two) on it with a simple A/B/Both switch.

  8. Meh… whatever!

    Bring me a 19″ rack with 4 Selectable Midi In, and 32 Midi Out that I can setup the way I want, directly in USB from my computer, have any kind of routing from one midi port channel to any other port channel, and such,… then it would be a great “Midi Patchbay”.

  9. The lack of USB does not deter me from buying it – quite the contrary. This box will work long after you next upgrade your DAW, or OS without worrying about driver compatibility etc. as it’s a raw midi box. I hate the hassle of upgrading drivers etc. each time I upgrade my OS, or trawling forums to work out why something no longer works. Maybe I am just getting lazy in my old age 🙂

Leave a Reply