LSTM - unable to get a 3D output

I have an array with shape (55834, 250, 30) and I'd like to get an output of the same shape from this LSTM model.

self.model = Sequential()

self.model.add(LSTM(
    self.config.layers[0],
    input_shape=(channel.X_train.shape[1], channel.X_train.shape[2]),
    return_sequences=True))
self.model.add(Dropout(self.config.dropout))

self.model.add(TimeDistributed(Dense(
    self.config.n_predictions)))

self.model.add(Activation('linear'))

self.model.compile(loss=self.config.loss_metric,
    optimizer=self.config.optimizer, metrics=[mse])

self.model.fit(channel.X_train,
    channel.y_train,
    batch_size=self.config.lstm_batch_size,
    epochs=self.config.epochs,
    validation_split=self.config.validation_split,
    callbacks=cbs,
    verbose=True)

If I run it I get the error:

ValueError: Error when checking target: expected activation_1 to have 3 dimensions, but got array with shape (55834, 30)

Why am I losing one dimension? Doesn't the TimeDistributed layer recreate a 3D tensor?