Mycelium Android derivation path (for eth)

[kstunna86]

This is related to a stuck tx issue I posted about that got moved to a different forum so I will make this post more specific to the Mycelium wallet.

I am trying to confirm the derivation path that mycelium uses for getting the private keys for ethereum wallets, and more specifically when you have created multiple ethereum wallets.

When I try and use a standard derivation path on metamask or MEW only the very first eth address matches, and the rest of them do not.

I have also tried the iancoleman tool as well.

Thank you for any help or anyone who can point me in the right direction.

[BitMaxz]

Have you read my previous post in your previous thread?
I suggested to try to use https://iancoleman.io/bip39/ and then download it to run it offline.
The tool supports ETH you can paste the seed backup from Mycelium and you should see all private keys of your seed backup that you can use to import to Metamask.

[kstunna86]

I did, I am struggling with that Iancoleman tool because I dont know what or how to enter the derivation key for Mycelium.  The preset options do not generate the same tree of eth addresses.  Only the first one comes up the same.

I did some digging and it seems like mycelium for android opted for some sort of "mixed derivation" key back when segwit was introduced?

[Rickorick]

Your best bet is iancoleman. you have to tweak the paths till you're able to find the missing address.'

[nc50lc]

I did, I am struggling with that Iancoleman tool because I dont know what or how to enter the derivation key for Mycelium.

Try turning-on the setting "Show address path".
With it, it'll show you the address' derivation path in the 'balance' tab.

-snip-, and more specifically when you have created multiple ethereum wallets.

As for ETH accounts in Mycelium's 'Account' tab, each "Ethereum N" account corresponds to another account_index derivation, not a higher address_index.

For example:
If 'Ethereum 1' has the derivation path of m/44'/60'/0'/0/0, 'Ethereum 2' should be m/44'/60'/1'/0/0
In IanColeman's BIP39 tool, just change "Account" from 0 to 1 to get Ethereum 2's prvKey.

[libert19]

I did, I am struggling with that Iancoleman tool because I dont know what or how to enter the derivation key for Mycelium.

Try turning-on the setting "Show address path".
With it, it'll show you the address' derivation path in the 'balance' tab.

-snip-, and more specifically when you have created multiple ethereum wallets.

As for ETH accounts in Mycelium's 'Account' tab, each "Ethereum N" account corresponds to another account_index derivation, not a higher address_index.

For example:
If 'Ethereum 1' has the derivation path of m/44'/60'/0'/0/0, 'Ethereum 2' should be m/44'/60'/1'/0/0
In IanColeman's BIP39 tool, just change "Account" from 0 to 1 to get Ethereum 2's prvKey.

I replied op on his another thread where he was specifically asking to get pvk for 23rd derived address, and I'm pretty sure this is correct path for op's 23rd address — m/44'/60'/22'/0 — I tried it practically on my mycellium to fetch mine, so I'm pretty sure, however it'd have been good if op himself could confirm same but he's not been active.

Only difference between ours is one less zero, idk if it makes a difference.

Ok thank you for highlighting that setting. Didn't know that was in there.  It seems that my derivation path for the specific wallet is:

m/44'/60'/22'/0/0

or the very first eth address I created is m/44'/60'/0'/0/0

Using the 44/60/0/0/0 one again only nets me the first correct address, and all the others are different.

Ok, do this. Go on MEW wallet website, import your mnemonic, add this derivation path without quotes, "m/44'/60'/22'/0" — select it and the first address you will see there will of your mycellium wallet.

[nc50lc]

-snip-

I replied op on his another thread where he was specifically asking to get pvk for 23rd derived address, and I'm pretty sure this is correct path for op's 23rd address — m/44'/60'/22'/0 — I tried it practically on my mycellium to fetch mine, so I'm pretty sure, however it'd have been good if op himself could confirm same but he's not been active.

Only difference between ours is one less zero, idk if it makes a difference.

Not much of a difference if the reader knows the BIP44 standard which describes each level of the derivation path.
The extra zero is for the 'address_index' if going from seed to private key/address.
Link: github.com/bitcoin/bips/blob/master/bip-0044.mediawiki#user-content-Path_levels

This is quite technical for newbies so I've just instructed him to change the "account" in BIP39 tool accordingly where he already successfully derived the address of "Ethereum 1".

The path you've given him should be correct without a doubt if it's his "Ethereum 23" wallet.
because as I've mentioned, each of those ETH wallets in the 'accounts' tab are derived with successive 'account_index'.